Thanks for engaging Jeff. A beer would be nice. Your photo is Hong Kong, no? I'll be there in Dec if you are! We can see if density there has made homes cheap ;-)
On the topic at hand, what's you view on why homes are built? Why are they not built? That's the key to all this.
Once we get past our areas of agreement, it seems to boil down to these questions.
I sense your answer is that property owners are in a hurry to build more homes (more per period, not just more density on a site) regardless of demand conditions. That as soon as redevelopment has a higher residual than the current use, property owners will rush to redevelop without regard to their effects of those actions on soaking up demand and hence changing future rents and prices. In this view, "supply" (rate of new housing produced in a period of time across a region) is tightly controlled by regulations, not by market choices of property owners about when to redevelop. But planning regulations don't control WHEN property owners choose to develop, only WHERE.
I disagree. And the economics on this is pretty clear, just widely ignored, as I explain here
What is more interesting is when you look around the world and historically you find the property market doing the same old thing.
Your argument seems to revolve around wanting a different spatial distribution of dwellings. Fine. Planning DOES change the spatial distribution of homes, as intended. But make the spatial distribution more efficient will make average dwellings better and hence higher value.
Mike Fellman explains this in his Boyd institute interview.
Anyway, I have a book digging into the many details in this debate, so perhaps we should catch up after you've read that.
Hey Cameron, thx for the read. And good eye, that photo is totally in HK, though I'm presently living in Seattle, so you may have to make it to the west coast in order to claim that beer :)
So, I feel like we may be talking past each other. If I'm understanding you correctly, it seem that you believe:
- the total quantity of housing supplied during some time period, n, is dictated solely by aggregate demand
- and also that making land use restrictions more/less permissive just changes the location of the housing built without having any impact on the total quantity produced during n
Hopefully I'm tracking so far. I'm also reading you as saying:
- landowners (I'm interpreting this as potentially distinct from developers) will wait to build if they believe future returns from land appreciation make it relatively more profitable to do so.
Fwiw, I mostly agree with this part. I don't think landowners necessarily press play on build the moment development has a higher residual than the current use (particularly if they believe they're no where near the top of the market cycle). That said, you also seem to believe that:
- landowners effectively coordinate their build decisions to keep prices high
This would seem to imply well-coordinated cartel behavior? Alternatively, perhaps you're actually claiming:
- capital markets are coordinating landowner behavior such that they're acting as if they were in a cartel
If this is the claim, it would seem to imply a) efficient capital markets, b) financial capital as a homogenous stock c) homogenous time preference among landowners?
Let me know what I'm not getting. Taking a different tack, it might be useful to turn to empirics for a second.
I used the Houston example in my post because the '98 reform is one of the most aggressive land use liberalizations in the history of the United States. And, as a brief aside, Houston's land use is still quite regulated in a number of ways, they just don't use Euclidean Zoning or discretionary permitting.
So the reform reduced minimum lot sizes from 5,000 sqft to 1,400 sqft for townhomes in the inner loop (a 96 square mile area). This kicked off a building boom, most of which was priced competitively for middle income Houstonians. New construction providing housing products for the middle of the income distribution would seem to cut against your description of housing markets, no?
Putting that aside, would you look at something like Houston's liberalization and still say:
- all the construction that happened post reform was simply reallocated from...other places in the city's housing market? The Texas housing market? The U.S. national housing market? And if we're sticking with a story about spatial reallocation, again, are we assuming capital markets are the coordinating mechanism?
Houston? Yeah, probably mostly a reallocation with the city, and nationally a reallocation for sure.
I wouldn't say "capital markets are the coordinating mechanism". Just markets. People move, trade, build, wait, etc. These are all market trades. None has to be perfect for the trading system to generate predictable emergent outcomes.
Making distribution more efficient doesn’t mean making dwellings better or higher value. "Better" doesn’t necessarily mean better suited. If people are forced to consume more land than is efficient, they consume more land and less of everything else.
By your logic, forcing people to eat twice as much would somehow lower food bills. Restricting economising makes things less economical. Banning smaller, cheaper, or shared options like rooming houses or hostels makes the whole system less efficient and hits the poorest hardest.
Existing regulation definitely does not stop speculation or get homes built now. It creates even greater speculation on exceptions. The option value of land rises with volatility, and there’s far more volatility in lobbying to turn five detached houses into five hundred apartment units, a change that could only occur because decades of binding supply constraints made it possible, than in the typical mere 3% annual market swing.
Your argument is that regulations get homes built today. Regulation like an outright ban clearly wouldn't get homes built today. We're closer to that than the free-market scenario where a pinch of regulation could reduce volatility.
As a transportation planner, I assure that the spatial distribution of buildings is extremely important from a transportation perspective. And that uniform density generates pretty much the worse possible outcomes in terms of maximizing required travel distances, limiting labor market access by non-car owners, making effective transit infeasible, losing time in traffic, and suffering air pollution.
I work for large, international home builders. If we do not have clear permission, or as we would say in the states “entitlement“ to construct, then we cannot expect a rational investor to support a project without commensurate reward.
Capricious land use policy, and the ability for community groups to delay or cancel projects, increases the cost of building by increasing the price of investment, and by extending the investment horizon. There is a real cost to tying capital down for 10 years on what should be a five year project.
So, no surprise that large builders simply avoid regions that are Nimby by policy or proclivity.
As for government subsidized housing… You do know we have tried this before? The results were literally ugly.
It turns out that the government is really good at slowly building, dysfunctional, ugly, unpopular ineffective… Well, I seem to be running out of adjectives.
Anyway, the underlying difficulty is the US love affair with single-family housing. Even when the houses are built zero lot line, no lawn. We need a multi family product that appeals so strongly, that the average citizen is willing to put up with proximity to people. Nobody likes people.
Agreed on the love affair with SFH, but I don't think it's much of a tough sell.
Where we allow density, it gets built. I know a few folks jumping into the market for ADUs and missing middle as we're clearing away the restrictions. A more definitive example is probably the townhome construction boom in Houston following the 1998 minimum lot size reductions. People were have been very happy to embrace relative density in exchange for lower cost housing.
Speaking from my own experience, the residential developments I’m involved in are approximately 15% to 20% townhomes in terms of footprint. This is a typical nod to the desire for “moderately priced dwelling units” that I have seen over the last 40 years in the US mid Atlantic and Southeast.
I don't love living with the smells and sounds of my apartment neighbors. But I consider it an acceptable trade-off to being able to not have to drive and/or suffer traffic anyplace I want to go somewhere.
First, there is no "housing shortage." San Francisco reportedly has five times more vacant homes than its homeless population. The US has more vacant homes than its homeless population--the largest since the Great Depression. The problem is distribution.
Regarding the fate of subsidies, it's essential to recall that Nixon halted federal funding for affordable housing in the 1970s. As he reduced taxes on the wealthy by roughly half, Reagan slashed HUD's affordable housing budget by 75% (and his successor subsequently increased payroll taxes eightfold). Gee, I wonder why affordable housing is so scarce, and incomes are so unequal. What could the problem be?
As for whether the "over-regulation" of building matters, that's a fantasy. The regulations are a joke, changeable at the drop of a hat by whoever wields the most political influence. They're a convenient excuse for local governments, but simply not the problem builders cry wolf about. Why else would we have so many homes in, e.g., a floodplain? (See "North Natomas" in Sacramento.)
This is a very old article (written and published in 1993), but it still applies to California today: https://itssimplerthanitlooks.blogspot.com/2018/01/placemaking-matters.html It explains the logic behind land speculation in particular. The reason to build in a flood plain: you can get the land cheaply, and get the government to shore up the levees. Profits are often 5,000% - 10,000%, after tax!
Finally, some economists (Steve Keen, Michael Hudson) say banks love high house prices because they can make bigger loans. It's not "form follows function," it's "form follows finance." Just try to borrow money to build a boarding house, for example. The zoning is incidental. Unelected bank underwriters are the people in charge. If you can't finance or insure the construction, it won't get built.
Many years ago, I worked on analyzing when real estate would redevelop, and the analysis simply ignored single family homes, because they rarely redevelop. As long as you have land to spare, you can simply add more building. And in many cases, adding another story is also trivial. So it's not until you run up against fire code for wooden structures (at five stories) and 100% lot coverage that demolition and replacement is a better alternative to a structural retrofit. And that's a greater level of density than 99% of America.
I much enjoyed the explainer on "Markets efficiently delay building feasible new homes", which does a nice job explaining the long duration of vacant lots in so many downtowns, and the presence of so many low-rent 'taxpayer' land uses in Manhattan. And as a landlord, I have some homes where location/land value would certainly support more intense uses (even under present regulation). It's easy to think that I could build an ADU and not affect my redevelopment potential, but in reality, that would bring in more rent, and I'd have even less incentive to redevelop to a more intense use (as that would require demolishing even more of my invested capital).
Thanks for engaging Jeff. A beer would be nice. Your photo is Hong Kong, no? I'll be there in Dec if you are! We can see if density there has made homes cheap ;-)
On the topic at hand, what's you view on why homes are built? Why are they not built? That's the key to all this.
Once we get past our areas of agreement, it seems to boil down to these questions.
I sense your answer is that property owners are in a hurry to build more homes (more per period, not just more density on a site) regardless of demand conditions. That as soon as redevelopment has a higher residual than the current use, property owners will rush to redevelop without regard to their effects of those actions on soaking up demand and hence changing future rents and prices. In this view, "supply" (rate of new housing produced in a period of time across a region) is tightly controlled by regulations, not by market choices of property owners about when to redevelop. But planning regulations don't control WHEN property owners choose to develop, only WHERE.
I disagree. And the economics on this is pretty clear, just widely ignored, as I explain here
https://www.fresheconomicthinking.com/p/explainer-markets-efficiently-delay?utm_source=publication-search
What is more interesting is when you look around the world and historically you find the property market doing the same old thing.
Your argument seems to revolve around wanting a different spatial distribution of dwellings. Fine. Planning DOES change the spatial distribution of homes, as intended. But make the spatial distribution more efficient will make average dwellings better and hence higher value.
Mike Fellman explains this in his Boyd institute interview.
Anyway, I have a book digging into the many details in this debate, so perhaps we should catch up after you've read that.
https://www.amazon.com.au/Great-Housing-Hijack-keeping-Australia/dp/176147085X/ref=mp_s_a_1_1?crid=3SE0YQSOAC9DD&dib=eyJ2IjoiMSJ9.KLD7-bxkI5wjl2TA5DgDyA.G2H1exl8zJbIqQhaguwScroB7jUXrXrzRF4GqTRbdHc&dib_tag=se&keywords=the+great+housing+hijack+by+cameron+murray&qid=1753139618&sprefix=great+housing+hijack,aps,307&sr=8-1
Hey Cameron, thx for the read. And good eye, that photo is totally in HK, though I'm presently living in Seattle, so you may have to make it to the west coast in order to claim that beer :)
So, I feel like we may be talking past each other. If I'm understanding you correctly, it seem that you believe:
- the total quantity of housing supplied during some time period, n, is dictated solely by aggregate demand
- and also that making land use restrictions more/less permissive just changes the location of the housing built without having any impact on the total quantity produced during n
Hopefully I'm tracking so far. I'm also reading you as saying:
- landowners (I'm interpreting this as potentially distinct from developers) will wait to build if they believe future returns from land appreciation make it relatively more profitable to do so.
Fwiw, I mostly agree with this part. I don't think landowners necessarily press play on build the moment development has a higher residual than the current use (particularly if they believe they're no where near the top of the market cycle). That said, you also seem to believe that:
- landowners effectively coordinate their build decisions to keep prices high
This would seem to imply well-coordinated cartel behavior? Alternatively, perhaps you're actually claiming:
- capital markets are coordinating landowner behavior such that they're acting as if they were in a cartel
If this is the claim, it would seem to imply a) efficient capital markets, b) financial capital as a homogenous stock c) homogenous time preference among landowners?
Let me know what I'm not getting. Taking a different tack, it might be useful to turn to empirics for a second.
I used the Houston example in my post because the '98 reform is one of the most aggressive land use liberalizations in the history of the United States. And, as a brief aside, Houston's land use is still quite regulated in a number of ways, they just don't use Euclidean Zoning or discretionary permitting.
So the reform reduced minimum lot sizes from 5,000 sqft to 1,400 sqft for townhomes in the inner loop (a 96 square mile area). This kicked off a building boom, most of which was priced competitively for middle income Houstonians. New construction providing housing products for the middle of the income distribution would seem to cut against your description of housing markets, no?
Putting that aside, would you look at something like Houston's liberalization and still say:
- all the construction that happened post reform was simply reallocated from...other places in the city's housing market? The Texas housing market? The U.S. national housing market? And if we're sticking with a story about spatial reallocation, again, are we assuming capital markets are the coordinating mechanism?
Anyway, thanks for the back/forth.
Cartel behaviour is not necessary. It is normal competition in the property market over time that comes from the basic ideas in Hotelling from 1931. https://www.fresheconomicthinking.com/p/crampton-debates-crampton-on-the?utm_source=publication-search
Houston? Yeah, probably mostly a reallocation with the city, and nationally a reallocation for sure.
I wouldn't say "capital markets are the coordinating mechanism". Just markets. People move, trade, build, wait, etc. These are all market trades. None has to be perfect for the trading system to generate predictable emergent outcomes.
Making distribution more efficient doesn’t mean making dwellings better or higher value. "Better" doesn’t necessarily mean better suited. If people are forced to consume more land than is efficient, they consume more land and less of everything else.
By your logic, forcing people to eat twice as much would somehow lower food bills. Restricting economising makes things less economical. Banning smaller, cheaper, or shared options like rooming houses or hostels makes the whole system less efficient and hits the poorest hardest.
Existing regulation definitely does not stop speculation or get homes built now. It creates even greater speculation on exceptions. The option value of land rises with volatility, and there’s far more volatility in lobbying to turn five detached houses into five hundred apartment units, a change that could only occur because decades of binding supply constraints made it possible, than in the typical mere 3% annual market swing.
Your argument is that regulations get homes built today. Regulation like an outright ban clearly wouldn't get homes built today. We're closer to that than the free-market scenario where a pinch of regulation could reduce volatility.
As a transportation planner, I assure that the spatial distribution of buildings is extremely important from a transportation perspective. And that uniform density generates pretty much the worse possible outcomes in terms of maximizing required travel distances, limiting labor market access by non-car owners, making effective transit infeasible, losing time in traffic, and suffering air pollution.
I work for large, international home builders. If we do not have clear permission, or as we would say in the states “entitlement“ to construct, then we cannot expect a rational investor to support a project without commensurate reward.
Capricious land use policy, and the ability for community groups to delay or cancel projects, increases the cost of building by increasing the price of investment, and by extending the investment horizon. There is a real cost to tying capital down for 10 years on what should be a five year project.
So, no surprise that large builders simply avoid regions that are Nimby by policy or proclivity.
As for government subsidized housing… You do know we have tried this before? The results were literally ugly.
It turns out that the government is really good at slowly building, dysfunctional, ugly, unpopular ineffective… Well, I seem to be running out of adjectives.
Anyway, the underlying difficulty is the US love affair with single-family housing. Even when the houses are built zero lot line, no lawn. We need a multi family product that appeals so strongly, that the average citizen is willing to put up with proximity to people. Nobody likes people.
Agreed on the love affair with SFH, but I don't think it's much of a tough sell.
Where we allow density, it gets built. I know a few folks jumping into the market for ADUs and missing middle as we're clearing away the restrictions. A more definitive example is probably the townhome construction boom in Houston following the 1998 minimum lot size reductions. People were have been very happy to embrace relative density in exchange for lower cost housing.
Speaking from my own experience, the residential developments I’m involved in are approximately 15% to 20% townhomes in terms of footprint. This is a typical nod to the desire for “moderately priced dwelling units” that I have seen over the last 40 years in the US mid Atlantic and Southeast.
I don't love living with the smells and sounds of my apartment neighbors. But I consider it an acceptable trade-off to being able to not have to drive and/or suffer traffic anyplace I want to go somewhere.
First, there is no "housing shortage." San Francisco reportedly has five times more vacant homes than its homeless population. The US has more vacant homes than its homeless population--the largest since the Great Depression. The problem is distribution.
I've recently read some hand-wringing about how affordable housing doesn't make unsubsidized economic sense. (https://www.nakedcapitalism.com/2025/11/losing-money-every-month-growing-finance-crisis-threatens-affordable-housing-challenges-mamdani.html)
Regarding the fate of subsidies, it's essential to recall that Nixon halted federal funding for affordable housing in the 1970s. As he reduced taxes on the wealthy by roughly half, Reagan slashed HUD's affordable housing budget by 75% (and his successor subsequently increased payroll taxes eightfold). Gee, I wonder why affordable housing is so scarce, and incomes are so unequal. What could the problem be?
As for whether the "over-regulation" of building matters, that's a fantasy. The regulations are a joke, changeable at the drop of a hat by whoever wields the most political influence. They're a convenient excuse for local governments, but simply not the problem builders cry wolf about. Why else would we have so many homes in, e.g., a floodplain? (See "North Natomas" in Sacramento.)
This is a very old article (written and published in 1993), but it still applies to California today: https://itssimplerthanitlooks.blogspot.com/2018/01/placemaking-matters.html It explains the logic behind land speculation in particular. The reason to build in a flood plain: you can get the land cheaply, and get the government to shore up the levees. Profits are often 5,000% - 10,000%, after tax!
Finally, some economists (Steve Keen, Michael Hudson) say banks love high house prices because they can make bigger loans. It's not "form follows function," it's "form follows finance." Just try to borrow money to build a boarding house, for example. The zoning is incidental. Unelected bank underwriters are the people in charge. If you can't finance or insure the construction, it won't get built.
Many years ago, I worked on analyzing when real estate would redevelop, and the analysis simply ignored single family homes, because they rarely redevelop. As long as you have land to spare, you can simply add more building. And in many cases, adding another story is also trivial. So it's not until you run up against fire code for wooden structures (at five stories) and 100% lot coverage that demolition and replacement is a better alternative to a structural retrofit. And that's a greater level of density than 99% of America.
I much enjoyed the explainer on "Markets efficiently delay building feasible new homes", which does a nice job explaining the long duration of vacant lots in so many downtowns, and the presence of so many low-rent 'taxpayer' land uses in Manhattan. And as a landlord, I have some homes where location/land value would certainly support more intense uses (even under present regulation). It's easy to think that I could build an ADU and not affect my redevelopment potential, but in reality, that would bring in more rent, and I'd have even less incentive to redevelop to a more intense use (as that would require demolishing even more of my invested capital).
Did you submit an essay to the Boyd contest?
I did, you?
Yes. I think you have a good chance of winning judging from your past writing quality and scope of ideas. Good luck
Thank you, I appreciate that — good luck to you, too