252 Comments
User's avatar
PJC's avatar

If you build it, they will come.

Does Supply Create Its Own Demand?

https://www.tenantstogether.org/updates/does-supply-create-its-own-demand

This guy gives a similar explanation to mine. He's a planner. Not an academic or activist.

No, new MR housing construction doesn't technically "create" demand, it simply satisfies "latent" demand. My explanation is more causal: new office construction INDUCES new local housing demand by attracting in-migration. "Static" housing supply models are incorrect. They do not account for the impact of new housing on new office (local demand) and they do not account for "geo-spatial" mobility both by idle world elites and from future tech workers.

Money quotes:

"The problem is that under some market conditions, more supply doesn’t lead to market equilibrium because it actually creates its own demand. You can see this wherever the world’s uber-rich decide to buy houses – New York, London, or, most relevant to this discussion, Santa Barbara.

...

Santa Barbara has housing prices that are not supported by the underlying dynamics of the local economy, for one very simple reason: The uber-rich from around the world drive up home prices by paying premium prices, often for houses they don’t actually occupy very often ...

[PJC: he's describing "latent" demand not "induced" demand. And yes, we see "ghost" houses all over Silicon Valley, and local realtors actively market in China.]"

...

"Something similar is going on today in San Francisco and Santa Monica. These places are hotbeds for cool jobs. The folks taking the cool jobs may not be uber-rich, but they have tons more money than everybody else, and so they drive prices out of sight. Build more market-rate housing, and you’ll just accelerate the cycle – more smart kids will show up wanting to work for tech start-ups, and that means you’ll have more tech start-ups, and pretty soon demand will rise faster than supply – in large part because you increased the supply. To a local community activist, it feels like a no-win."

[he's conflating "latent" and "induced" demand and is attributing it to housing not office.]

IMO, he has the ordering wrong. FIRST offices are built, then "smart kids" show up to take the jobs in the offices. "Smart kids" usually don't just move here and then look for a job, though they may.

Tech firms continue to want to move to Silicon Valley because of so-called "agglomeration." Lack of housing for future employees might delay their decision or cause them to choose Austin instead, but new supply of housing would accelerate their decision to move to SV. Decisions are probably made by considering known housing price signals. High housing prices might be acting as a barrier to new job formation in SV, and that is truly what local business groups are worried about.

PJC's avatar

You're right to criticize a wrong theory of "induced demand" but you're wrong to turn it into a straw man. And, there is a real, correct theory of induced demand.

Both "latent" demand and "induced" demand are real economic concepts used frequently in traffic projections. Latent demand is existing, current demand that would consume supply if supply existed. "Induced" demand is not-yet-existing demand that will exist once a demand-inducing trigger event occurs. Both may be called "pent-up" demand and are easy to conflate.

Whether or not they apply to any particular housing sub-market, both do occur near where I live and was once an elected official and mayor in Menlo Park, California. (For reasons beyond the scope of this post, "filtering" is insignificant.)

INDUCED growth is prevalent. The trigger is OFFICE CONSTRUCTION and resulting demand not HOUSING construction. For example, in regional equilibrium, a new office building in Menlo Park will INDUCE housing demand by attracting new people into the area to fill the jobs housed by the new office, (or fill other jobs vacated via vacancy-chains of locals who abandon other local jobs to get the new ones.)

Here, new in-migrants fill jobs that usually pay higher than the median. Low productivity companies cannot afford to move here. If there is not enough offsetting new housing, the new demand will crowd out existing lower income residents. (In SM County a "low" income for a single person is $104k. The crowding-out impacts "richer" people than you might think, and this helps explain why "out-migrants" from California also have seemingly high incomes.)

Induced growth and displacement are normal assumptions in official legal analyses. For example, consider the Housing Needs Assessment (HNA) for the recently adopted Facebook (Meta) willow village project in which FB got approvals to build >1M sf of new office, and hotel along with 1700 new housing units. The HNA estimates that despite the new housing the project WORSENS the housing deficit to the tune of DISPLACING 815 low income families somewhere in San Mateo County. The project creates a slight housing surplus of higher income units but a net total housing deficit. The HNA also has a nice empirical discussion of local rents and median incomes that speak to the issue of affordability and whether or not market rate housing is "affordable" and to whom.

IMO, the Willow Village project is a representative anecdote. Proving it extrapolates is beyond the scope of my comment. Here, I simply assert that office construction outpaces and crowd-outs housing construction. New market rate housing is affordable only to "above median" incomes. You can pretend to remove as many alleged regulatory obstacles to supply as you want and this won't change.

I know you don't want to hear this, but its more a market-failure (or byproduct) than a regulatory failure.

LATENT demand is less significant. It consists of affluent non-locals or corporations that would rent or buy units in, say, downtown Menlo Park or Palo Alto if more existed. It's hard to measure https://www.paloaltoonline.com/real-estate/2015/04/24/are-ghost-homes-becoming-a-problem/ The most significant point in this link is that local realtors actively market and sell Bay Area houses into select Chinese markets using an array of on-line tools. If you look at 10 year census results of the same Bay Area towns you'll see large, disproportionate percentages of homes owned by self-describe Asians. Census can't tell us what percentage are foreign nationals.

The point: Local market actors are actively cultivating remote buyers.

Finally, induced housing is official assumed by every affordable housing impact fee Nexus study done in any city in California to comply with its law AB1600. These studies lay out the number of induced jobs created by new construction, commercial AND residential, the income levels of the jobs created, and the amount of subsidization (impact fee) required to house the induced jobs that pay lower income in the local markets (at median prices). Cities usually only assess a fraction of the full market cost.

Go to the bottom of the page here https://www.smcgov.org/planning/affordable-housing-impact-fee-0 to see both residential and commercial nexus studies for San Mateo County. You can read them to deconstruction some of the economic logic used.

So the official analysis does assume some level of induced growth even from new residential construction, even though office construction is the real driver.

Miles's avatar

Might I suggest the observation of "induced demand" is actually detecting something else, which is "where are there local boards that are allowing developments"? I suspect that is why it's never just one new building that goes up - because many developers want to be building in attractive neighborhoods, but many planning boards are stopping them.

But once you see a big building headed up, you've observed a location where the NIMBYs have lost control of the planning board. And if the NIMBYs aren't calling the shots, many projects can get going...

Zachary Ethan's avatar

I'm not an economist so I don't have the correct terminology to make this argument, but here goes anyway:

It seems to me that studying the effects of isolated developments on prices within a small area understates the effect that a disbursed city-wide construction program would have since doing tons of new construction in a single area will concentrate the new wealthier people whereas a distributed program would spread them out so their gentrifying dollars would be spread across more areas?

I bring this up because I really don't understand why people on the left can't get behind the framing of "we need to build new housing everywhere - but mostly in rich areas." The problem of these intense displacements happen precisely within the existing policy framework where we pick some poor areas to spruce up, flood them with capital, and push residents out to poor suburbs. Instead we could build lots of housing everywhere and these mass displacements wouldn't happen!

Brian Bulger's avatar

It seems to me that far fewer people should own a home than currently do. People have a tremendous amount of wealth tied up in their house which puts them in a position of wanting to protect their house price. But everyone protecting their house price makes housing supply a problem.

Edward's avatar

Is anti-gentrification just a form of woke-redlining?

Edward's avatar

Pragmatic liberalism. Thanks. Gentrification also creates more wealth in minority communities (probably just a few people) that own property that is purchased by Developers at a premium. For those concerned with wealth inequities...this is how wealth is created. The children of orange farmers in Florida are now wealthy land barons!

homechef's avatar

The induced demand argument deserves treatment similar to the inflation skeptic argument. It's wrong in aggregate, but there is an insight to be had there. From inflation skeptics Matt gleaned the brilliant insight that the things where Baumol's cost disease bites the hardest are also the pieces that are particularly important (eg. childcare, schooling, medicine, food).

I think the insight from the induced demand case is that if I'm someone who is in the process of being priced out of a gentrifying market, it's likely that the YIMBY solution actually won't do anything for me. The issue isn't one of distance but segmentation. Here's the sketch of my argument:

1. actually affordable housing is not the result of government rules or what the housing was made as, it's the result of some *impairment* - nextdoor to a superfund site, in a "bad" neighborhood, with poor train connections, etc.

2. people who live in this neighborhood have traded off cheaper housing for that impairment, which they have selected into (it's a crappy neighborhood, but they have figured out how to do their shopping, they keep an eye out for their neighbors, etc).

3. improving the neighborhood - adding housing - especially nice housing, adding chain stores, pretty much any improvement, really - means that that tradeoff is gone.

4. I don't doubt that this is better for everyone else who moves in. But it's shitty for me. I lose my "investment", and pretty soon I lose my home. Now, I need to not only find another "affordable" place, I need to also "invest" in figuring out a new crappy neighborhood, etc.

5. The people who benefit are not here, the people who lose are here, so in effect we're going to fight to save the *impairment* of the current place! Since we're here and they aren't, we'll complain about developers, environmental impact, whatever.

In theory, since so many people are going to gain from my loss, there should be a deal to be made. What's that deal?

Augusta Fells's avatar

I think the trouble here is that it's pretty clear that (1) is not backed up by evidence... Or at least that there is quite a bit of evidence to the contrary. Looking at places that don't have a housing crisis, or even cities that do but *used" to allow substantial new housing (NYC through the 50s was desirable *and* broadly affordable to the middle class) it's easy to find examples of affordable housing without a superfund site. We've just had such bag housing policies in so many places since the 70s that it's hard to see that!

Cascadian's avatar

"... a Giant, ... a Wawa, ... and a bunch of other national chains along with some locally owned businesses."

Giant and Wawa seem closer to locally owned than to national: they're in a handful of states in one region of the country. I suspect almost no one within 1500 miles of here has heard of either.

Kareem's avatar

Both are regional, but of the two only Wawa is local (its HQ is in suburban Philadelphia). Giant is owned by the Dutch company Ahold Delhaize; its American holdings are concentrated on the East Coast (it holds Giant, Stop & Shop, and Food Lion among others) but it has chains scattered across continental Europe (plus one chain in Indonesia, probably because that's a former Dutch colony).

Kareem's avatar

And by "local" I mean "headquartered in the U.S. region it serves."

Trevor Ewen's avatar

I've decided this left anti-gentrification argument has an evil-twin in the right-wing zone of discourse.

Immigration: People on the right are concerned about crimes that immigrants are less likely to commit than native born residents. They are obsessed with labor competition that is not borne out in the evidence. They fear the eventual (leftward) political harms that immigrants will produce. Instead, a fair number of immigrants beat expectations and are more right wing or engage in lower political activity. It's a persistent, powerful narrative that is devoid of demonstration, and is entirely based on false assertions.

The gentrification argument has a similar feeling to me. "I don't care what the literature, the academics, the studies, and basic supply / demand logic says! I feel like something bad is happening."

Right wingers own the immigration discussion. Activists own the gentrification discussion. They both get mired in the unprovable, quasi-religious assertions that undergird their only means of continued relevance.

Chris's avatar

“You don’t need a “developer” to convert your garage into an ADU, you need a contractor. And if you drive out to a rural area, there’s no such thing as a “developer;” there are just housebuilders.”

How would Matt and others on this thread define “developer”? I find it genuinely confusing when I hear people say there shouldn’t be developers or there wouldn’t be developers if this or that were different. To me a developer performs functions related to overall project coordination and risk that are necessary and unavoidable. Of course, I’m biased. I am a developer!

To use Matt’s example, for a homeowner to build an ADU they need to determine whether likely rents justify the cost to build, acquire financing, come up with a design, oversee a contractor and take on the risk that the whole thing works out. Matt is probably right, the homeowner doesn't need another person to be the developer. I argue the homeowner is the developer.

Doctor Memory's avatar

At this point, I despair of any progress being made on this argument, at least in our major cities. The only YIMBY victory of note in a place with a real housing crisis so far is California's ADU law, and I think they only got away with that because converting a garage into an in-law apartment involves nothing being built _upwards_ and in fact often no visible construction taking place at all.

Meanwhile I keep having conversations with my neighbors in which it's... not even "asserted", it's assumed to be a simple fact of nature that tall buildings are bad, and that we should at all costs prevent anyone from building anything taller than the building it's next to. This would merely be depressing except that I live in _Manhattan_.

I think the great untold story of the 21st century is the tacit agreement on both the left and the right that the ideal and just social world is one where nobody ever moves to a new country, state, city or even neighborhood, that doing so is an action at best deeply suspicious if not actively hostile, and that anyone who attempts to make such things easier at any level is the enemy of all humanity.

Ben Supnik's avatar

With California's ADU law, the released financial benefit goes to the home-owner, right? E.g. if I have a house that's eligible, I can trade some space for rent, and maybe the house value goes up because a future buyer will have that income stream too?

Maylor's avatar

“We simply cannot accept “make sure the neighborhood sucks” as our affordable housing strategy. “

Years ago traveled from Seattle to Eugene, OR, to attend the Oregon Country Fair (yep, the 90s). The neighborhood where we were staying had trash and graffiti everywhere. We asked, ‘what’s up with the trash?’ Answer: ‘Oh, that’s what the anarchists do, trash the neighborhood to keep the rents down.’

In the PNW there is generational tradition of dirtbags self inflicting harm to slum it. It’s part of the struggle.

Doctor Memory's avatar

In the dot-com era, the "Mission Yuppie Eradication Front" in San Francisco regularly torched people's parked cars as a means of trying to convince people to not move to SF's southern neighborhoods.

Troy a Garrett's avatar

I am curious if you had an initiative in San Francisco that said for the next 5 years all new new construction is only subject to state and federal laws. Like what would happen

Owen T.L.'s avatar

As a leftist activist type myself, I hate to admit you're right about the "Don't make neighborhoods better!" philosophy, but this absolutely does exist. I've heard this argument used against parks and community gardens, and it just turns my stomach.

Marc Robbins's avatar

This is a very compelling post, but I think it addresses the wrong question (i.e., housing prices by itself). The question I would like to see addressed is that when housing stock in a poor area is improved by market-rate construction, what happens to current residents? Are they able to stay? Are they forced out? If they're forced out, where do they go? If they're forced out, is that a net benefit or net cost to them over time? How many owned and sell their properties at a profit, versus being renters who gain no benefit if they have to leave the neighborhood?

I'd be interested in links to the academic literature on this.

Peter G's avatar

This is susceptible to analysis. If you build a new high rise apartment then where do the tenants come from. The same area or maybe out of town completely? They will predominantly be local seeking to move upscale. And every single one of them, unless they were living on the streets, will be leaving unoccupied housing behind them. Maybe these new tenants are just seeking better amenities and a shorter commute. In which case I will point out that shorter commutes reduce demand on commuter systems. Which is also good.

Marc Robbins's avatar

Agreed. Still my question is not about the incoming new tenants but those who lived in places that were replaced with the new construction. I'm not making any value judgments about that, but just wanted to be better educated on what happens to them beyond the fact that they're not living in their old places any more. Do they wind up better off? Worse off? Facts would be useful.

Peter G's avatar

I suppose it depends on whether they own or rent. If you own then the benefits of rising house prices is immense. Anecdotally I saw this with my own eyes. Once upon a time a developer called Huang and Danczkay built a bunch of condos on some quite valuable real estate. They were right on the waterfront of Toronto south of the Gardiner Expressway. And found they almost couldn't give them away. Transportation sucked, there were few amenities and the highway blocked them from the rest of the city. Fast forward twenty years to the point where the Skydome was built and transportation extended and the same condo a teacher bought for 175,000 was suddenly worth over a million. Millions if it was a three bedroom.

I was involved in structural rehab then and worked on those buildings in the parkades. It was interesting seeing the Ferraris of the new tenants parked right beside the undistinguished Ford sedans of the old tenants.