Incrementalism is not a solution to Midtown's problem. If it were, there would already be incremental development there. The problem with Midtown is the greedy land speculators who can afford to hold because their tax appraisals are too low. This leaves us in bizarro world, where nobody builds there because the land is too valuable. LVT! Interestingly, in the late 1800s, a printer named JJ Pastoriza was buying up land in Midtown and built a shoddy cabin on one lot, which he donated to the local Georgist league that he founded.
Thanks again for your comment. I wasn't happy that this point didn't come out strong enough, so I edited the post slightly, quoted you to really emphasize this :)
I didn’t spell it out as explicitly as I should have, but this was part of my point. Any time you end up with a big obvious gap in present value and future value, speculators get into the market and lock it down. Absent a corrective market measure like LVT, the neighborhood gets stuck, exactly as you say, where nobody can build because the land is “too valuable.”
To take another example: the community I grew up in the homes are now regularly priced $3-5million for a simple ranch on <10k sf. What would the allowable density be to create incremental booming housing? To do apartment buildings we'd usually assume land price is ~10-20% of the total development cost. if you figure $500 per sf. then you'd get something like 50,000 square feet on the new lot. If you ruled zoning unconstitutional, that might happen--but under any politically feasible land-use regime no way. Anything less and it's worth more as a SF home than as a development site.
So what's the incremental alternative? ADU's, sure, add rental stock, but once you own a $5m house you're not likely to like rent an ADU on the open market for even $3k per month. It's for the nanny, or your inlaws twice a year, or your kid home from college.
Townhouses seem like another option. For sale units can probably command a higher land value per unit. Let's assume they are like 50% land (40% construction, 10% profit +/-) And assume that ~10k lot can build... 4-8 townhouses? The land price going in is $500k-$1M. And the final prices are $1M to $2M. Which if you did across like a lot of the bay area... you wouldn't even drive down the prices of those SF homes, b/c a) they are becoming rarer and b) they have their minimum price set by the redevelopment alternative.
Thanks Andrew - I think this is an important vein to engage with, as this problem appears all over the place.
One component I don't think you mentioned: some of those parking lots may be encumbered (by lease or easement) to nearby buildings, and therefore not even marketable. One would have to do some digging to identify what percentage of those properties are marketable.
Second, I think it would be really interesting if some historian are archivist were to better explore the process by which land transacted during these distributed incremental booms like NYC or Chicago especially. Like what really happened? Why weren't land speculators locking up land and waiting for prices to go up? Or if they were what stopped that from blocking the growth of the overall market. This seems important!
Because actually rapid incremental growth is actually uncommon in work history (not perhaps rare... but uncommon).
Yes exactly, I think we need to understand better both why it doesn't happen and, when it did happen, why?
My best guess about when incremental development happened rapidly at large scale, is that it was easy for existing owners to cash in themselves at the time. Ie, if the city had a street grid, and you were a farmer who owned a farm which was suddenly adjacent to the existing street grid...
- You didn't want to hold out, bc. you lived on and worked a farm that was about to be surrounded by development
- You could kind of see that the streetcar suburbs were going to just keep stretching past you, and it was mainly workforce housing, so, holding out wasn't going to make your land *that* much more valuable bc. there was going to be a lot of competing supply coming online
- You didn't need to be rich to subdivide and sell off your land, you probably just needed a surveyor to plat an extension of the grid for you. Maybe some cities would do that for you?
- You didn't need to fund a lot of infrastructure in advance -- most of those roads were dirt to start and the city improved them later. Did they have sewers right away, or did the sewer come in block by block? I don't know.
This seems plausible, but it's based by fragments of stories I know about, not systematic research.
These are possible stories, but I think it would be really helpful to find someone who does the careful work of exploring specific cases from the past to help create a general principle here. Because I don't know that we have the theory that explains these dynamics or how they respond to pressures, yet.
To clarify, are you making the claim here that a large scale commercial project actively retards development because it makes adjacent landholders behave more "speculatively" than they already were (e.g. prior to the next Texas donut or whatever)?
I’m saying that’s a risk, and if we want to see many future projects after the first one then we need to mitigate against that risk.
The bigger the gap between existing conditions and the next wave of development, the higher the risk.
For emphasis: I am not opposed to large-scale development. What I want to avoid is failure to bring sufficient supply to market, and what I’ve seen is that speculative deadlock prevents a lot of development even when regulatory barriers aren’t the obstacle.
Totally, I hear your point on not being against large scale development, per se. If I'm hearing you correctly, it's that you're against large scale development in situations or under conditions in which said development might actually impede additional development in it's immediate vicinity.
/If/ I'm following, does that imply the following scenarios?
(a) no development at all, hypothetical area continues to be undeveloped
-- b/c landowners don't see add'l development as economically viable
(b) one large scale development, everything else around it continues to be undeveloped
--b/c landholders believe there's some larger windfall if they hold out for some indeterminate amount of time
(c) small scale/next increment development, similar incremental development proliferates in the immediate area
--b/c landholders recognize some economically higher value use, but also don't believe there's some major upswing in land values that they can ride like a wave and pull the development trigger right at the crest?
And in the interest of over communication, I'm genuinely asking to better understand your argument; promise this isn't an attempt to adversarially interrogate the thesis.
BTW I didn’t mean to ignore your scenarios in my other comment, so to circle back:
Basically, yes I think your scenarios are high probability.
The nuance is with scenario B, how big does the first wave of new development need to be, relative to existing conditions, to trigger speculative gridlock? I don’t think we can know that.
Also, where is the line between B and C? I think it’s fuzzy and contextual - we can’t know that for sure either. (In Manhattan it’s clear that replacing midrise with highrise has not ground to a halt, that’s large scale but for the context it also seems incremental.)
So, I think your scenarios are basically right, but I’m trying to convey this is a probabilistic observation for us to grapple with, not a mechanical principle or law.
I think the question is: Given we want to bring a lot of housing units to market in a short time period, what is the most effective way to do that?
My observation is that in places where there aren't significant regulatory barriers, and where the market demand is clearly strong, large-scale development hasn't delivered as much as we'd expect.
Development in an area tends to *start,* and then not *finish,* because of speculative deadlock. "The land gets too expensive so nobody can build on it." I don't think it's a mechanical thing where a development "causes" the speculation, so much as a probabilistic thing, that once you get a few of these off the ground everyone internalizes that they can make a ton of money holding out for super-luxury high-rise land prices, and then development grinds to a halt because the land is too expensive to redevelop further.
So I'm making a few claims:
- The bigger the gap between existing conditions and the next wave of development, the higher the risk of speculative deadlock
- Conversely, the lower the gap between existing conditions and the next wave of development, the lower the risk of speculative deadlock
So, this post should not be taken as an argument against large scale development, but rather this observation that scaling that kind of development is hard. If we want to do large-scale projects at scale, removing the regulatory barrier won't be enough, we'll still have to find a way to keep the market liquid. And I'm not sure we know how to do that. LVT could help, and other things like taxes on vacant lots, etc. We'll have to experiment and try things and see if we can find what works.
In the mean time, I think we should focus more of our energy on incremental development, because in practice that's what we've seen successfully bring a lot of supply to market fast.
Incrementalism is not a solution to Midtown's problem. If it were, there would already be incremental development there. The problem with Midtown is the greedy land speculators who can afford to hold because their tax appraisals are too low. This leaves us in bizarro world, where nobody builds there because the land is too valuable. LVT! Interestingly, in the late 1800s, a printer named JJ Pastoriza was buying up land in Midtown and built a shoddy cabin on one lot, which he donated to the local Georgist league that he founded.
Thanks again for your comment. I wasn't happy that this point didn't come out strong enough, so I edited the post slightly, quoted you to really emphasize this :)
Cheers!
Yes.
I didn’t spell it out as explicitly as I should have, but this was part of my point. Any time you end up with a big obvious gap in present value and future value, speculators get into the market and lock it down. Absent a corrective market measure like LVT, the neighborhood gets stuck, exactly as you say, where nobody can build because the land is “too valuable.”
Thank you for the comment!
To take another example: the community I grew up in the homes are now regularly priced $3-5million for a simple ranch on <10k sf. What would the allowable density be to create incremental booming housing? To do apartment buildings we'd usually assume land price is ~10-20% of the total development cost. if you figure $500 per sf. then you'd get something like 50,000 square feet on the new lot. If you ruled zoning unconstitutional, that might happen--but under any politically feasible land-use regime no way. Anything less and it's worth more as a SF home than as a development site.
So what's the incremental alternative? ADU's, sure, add rental stock, but once you own a $5m house you're not likely to like rent an ADU on the open market for even $3k per month. It's for the nanny, or your inlaws twice a year, or your kid home from college.
Townhouses seem like another option. For sale units can probably command a higher land value per unit. Let's assume they are like 50% land (40% construction, 10% profit +/-) And assume that ~10k lot can build... 4-8 townhouses? The land price going in is $500k-$1M. And the final prices are $1M to $2M. Which if you did across like a lot of the bay area... you wouldn't even drive down the prices of those SF homes, b/c a) they are becoming rarer and b) they have their minimum price set by the redevelopment alternative.
Wicked problem
Thanks Andrew - I think this is an important vein to engage with, as this problem appears all over the place.
One component I don't think you mentioned: some of those parking lots may be encumbered (by lease or easement) to nearby buildings, and therefore not even marketable. One would have to do some digging to identify what percentage of those properties are marketable.
Second, I think it would be really interesting if some historian are archivist were to better explore the process by which land transacted during these distributed incremental booms like NYC or Chicago especially. Like what really happened? Why weren't land speculators locking up land and waiting for prices to go up? Or if they were what stopped that from blocking the growth of the overall market. This seems important!
Because actually rapid incremental growth is actually uncommon in work history (not perhaps rare... but uncommon).
Yes exactly, I think we need to understand better both why it doesn't happen and, when it did happen, why?
My best guess about when incremental development happened rapidly at large scale, is that it was easy for existing owners to cash in themselves at the time. Ie, if the city had a street grid, and you were a farmer who owned a farm which was suddenly adjacent to the existing street grid...
- You didn't want to hold out, bc. you lived on and worked a farm that was about to be surrounded by development
- You could kind of see that the streetcar suburbs were going to just keep stretching past you, and it was mainly workforce housing, so, holding out wasn't going to make your land *that* much more valuable bc. there was going to be a lot of competing supply coming online
- You didn't need to be rich to subdivide and sell off your land, you probably just needed a surveyor to plat an extension of the grid for you. Maybe some cities would do that for you?
- You didn't need to fund a lot of infrastructure in advance -- most of those roads were dirt to start and the city improved them later. Did they have sewers right away, or did the sewer come in block by block? I don't know.
This seems plausible, but it's based by fragments of stories I know about, not systematic research.
These are possible stories, but I think it would be really helpful to find someone who does the careful work of exploring specific cases from the past to help create a general principle here. Because I don't know that we have the theory that explains these dynamics or how they respond to pressures, yet.
To clarify, are you making the claim here that a large scale commercial project actively retards development because it makes adjacent landholders behave more "speculatively" than they already were (e.g. prior to the next Texas donut or whatever)?
I’m saying that’s a risk, and if we want to see many future projects after the first one then we need to mitigate against that risk.
The bigger the gap between existing conditions and the next wave of development, the higher the risk.
For emphasis: I am not opposed to large-scale development. What I want to avoid is failure to bring sufficient supply to market, and what I’ve seen is that speculative deadlock prevents a lot of development even when regulatory barriers aren’t the obstacle.
Totally, I hear your point on not being against large scale development, per se. If I'm hearing you correctly, it's that you're against large scale development in situations or under conditions in which said development might actually impede additional development in it's immediate vicinity.
/If/ I'm following, does that imply the following scenarios?
(a) no development at all, hypothetical area continues to be undeveloped
-- b/c landowners don't see add'l development as economically viable
(b) one large scale development, everything else around it continues to be undeveloped
--b/c landholders believe there's some larger windfall if they hold out for some indeterminate amount of time
(c) small scale/next increment development, similar incremental development proliferates in the immediate area
--b/c landholders recognize some economically higher value use, but also don't believe there's some major upswing in land values that they can ride like a wave and pull the development trigger right at the crest?
And in the interest of over communication, I'm genuinely asking to better understand your argument; promise this isn't an attempt to adversarially interrogate the thesis.
BTW I didn’t mean to ignore your scenarios in my other comment, so to circle back:
Basically, yes I think your scenarios are high probability.
The nuance is with scenario B, how big does the first wave of new development need to be, relative to existing conditions, to trigger speculative gridlock? I don’t think we can know that.
Also, where is the line between B and C? I think it’s fuzzy and contextual - we can’t know that for sure either. (In Manhattan it’s clear that replacing midrise with highrise has not ground to a halt, that’s large scale but for the context it also seems incremental.)
So, I think your scenarios are basically right, but I’m trying to convey this is a probabilistic observation for us to grapple with, not a mechanical principle or law.
I think the question is: Given we want to bring a lot of housing units to market in a short time period, what is the most effective way to do that?
My observation is that in places where there aren't significant regulatory barriers, and where the market demand is clearly strong, large-scale development hasn't delivered as much as we'd expect.
Development in an area tends to *start,* and then not *finish,* because of speculative deadlock. "The land gets too expensive so nobody can build on it." I don't think it's a mechanical thing where a development "causes" the speculation, so much as a probabilistic thing, that once you get a few of these off the ground everyone internalizes that they can make a ton of money holding out for super-luxury high-rise land prices, and then development grinds to a halt because the land is too expensive to redevelop further.
So I'm making a few claims:
- The bigger the gap between existing conditions and the next wave of development, the higher the risk of speculative deadlock
- Conversely, the lower the gap between existing conditions and the next wave of development, the lower the risk of speculative deadlock
So, this post should not be taken as an argument against large scale development, but rather this observation that scaling that kind of development is hard. If we want to do large-scale projects at scale, removing the regulatory barrier won't be enough, we'll still have to find a way to keep the market liquid. And I'm not sure we know how to do that. LVT could help, and other things like taxes on vacant lots, etc. We'll have to experiment and try things and see if we can find what works.
In the mean time, I think we should focus more of our energy on incremental development, because in practice that's what we've seen successfully bring a lot of supply to market fast.