2026-04-25_18-15-31
How can we solve the housing crisis?
There isn’t a single “fix”—housing crises are structural problems caused by a mismatch between supply, demand, financing, and regulation. You only get meaningful relief when multiple levers are pulled at once, and sustained over years. Here’s a grounded breakdown of what actually moves the needle.
Most shortages are fundamentally about supply constraints.
Constraint: Local opposition (“NIMBYism”) and planning bottlenecks often matter more than physical limits.
Even when zoning allows building, projects stall.
In places like Australia, development delays can add years—this directly inflates prices.
If building is expensive, supply won’t scale.
This is often overlooked: even with perfect zoning, high costs cap output.
Policy often accidentally pushes prices up.
The private market won’t house everyone.
Countries that maintain large public housing stocks tend to have more stable rents.
Even with more supply, renters need protection.
Done poorly, rent controls can reduce supply—so these need careful design.
Housing shouldn’t behave purely like a financial asset.
Pressure concentrates in a few cities.
Without this, demand keeps piling into the same locations.
The evidence is pretty consistent:
You can’t significantly lower housing costs without:
That’s why the problem persists—it’s politically difficult, not technically mysterious.
If you want, I can break this down specifically for Brisbane or Australia and point out which of these levers would have the biggest immediate impact locally.
2026-04-25_18-25-18
I think I have mentioned before that I write science fiction stories. In most of my stories I portray a future where people build down instead of up, letting the surface become parkland or even bushland once again.
It always seemed to me really stupid the way we take the most fertile and lush places, pave them over and build ugly, noisy cities and suburbia, often building with wood in a country notorious for its bushfires. When there is a fire, or other disaster, look where people shelter: in underground bunkers.
When I point out this obvious solution to most people they say "Ugh! I wouldn't like to live in a dark hole in the ground." But most underground homes are far better lit by natural and artificial lighting than almost all the above-ground houses I've been in. And people have no problem spending time in shopping malls which have no view of the outside world at all.
Underground homes offer safety from fire, earthquakes, war, thieves*, and much more. They're also cheaper to maintain at a comfortable temperature, and offer great sound insulation against noisy neighbors.
At the moment underground building can initially be more expensive than aboveground building, but heating and cooling an underground house is far less costly, so over the longer term underground housing would likely be much cheaper. Also, if underground housing became common, many of those initial costs would fall away.
One of the major forces against allowing large numbers of people to live close to a desirable location is, as you mentioned, NIMBYism (Not In My Back Yard). Underground building would largely solve that. Large numbers of people could live close to a small, pleasant town center without destroying its charm.
There are only two ways I can see for such a future to come about:
- government projects to build large numbers of homes for the homeless and the poor. The homes would have to be designed and built for luxury and efficiency, with all materials and labor sourced from Australia so that it in effect doesn't cost the government anything in the long term (money paid just returns to the government in tax), whereas labor and goods sourced from overseas is money lost (or potentially lost). If the homes are built for luxury, attractiveness, and efficiency, then they boost those industries and bring down prices and encourage investment in those things for all Australians. It is no good building ugly, inefficient, uncomfortable boxes that people don't want to live in and which become slums. That's been tried and it almost always works out badly.
- famous people, like movie stars see the wisdom of building underground and then most people, being imitative, would follow the fashion and we could quickly see much of ugly city and suburbia transformed into beautiful parkland.
* I mentioned safety against thieves. This is something most people don't consider. In an aboveground house there are many ways into the house, whereas an underground house can have its few entry points well armored. Also, if bad people do gain entry there is no way to know the extent of the home -- bad actors can't walk around the outside to see the building's limits. It can have large parts of the home hidden without obvious entry.
Excuse me. I didn't really intend for this to be a rant. :)
It’s not a rant—it’s a coherent design thesis. You’re basically proposing a subterranean urbanism model: shift the bulk of habitation below grade to reclaim surface land, stabilise climate loads, and decouple density from visual intrusion. That idea has real technical merit—but also some hard constraints that explain why it hasn’t scaled.
Let’s separate what works, what breaks, and what would need to change.
Below a certain depth, ground temperature is relatively constant year-round. That gives you:
This is essentially leveraging thermal mass at geological scale.
You’re absolutely right here—subsurface structures are naturally protected from:
Australia’s building codes already push toward “defendable space,” but underground largely sidesteps the problem.
Soil is an excellent sound attenuator. You eliminate:
That has measurable mental health benefits.
Your key insight: density without visual density.
You could have:
This directly neutralises one of the main drivers of NIMBY resistance.
Digging is expensive, full stop.
Even if lifecycle costs are lower, upfront capital kills adoption.
This is the engineering headache people underestimate.
This is solvable—but not cheaply or casually.
You already identified the core objection: perception.
Even if irrational, people strongly prefer:
This ties into environmental psychology. Lack of natural light and horizon cues can affect circadian rhythms and mood.
You can mitigate with:
…but perception still lags reality.
Fire safety flips in an interesting way:
Regulations require:
These add complexity and cost.
Cities aren’t just houses:
Subsurface urbanism needs system-level redesign, not just different houses.
Conceptually sound—but politically difficult.
You’re proposing:
This resembles a sovereign investment model, but governments typically operate on short electoral cycles.
Also, public housing historically fails when:
You’re right that quality is non-negotiable. Without it, it becomes vertical slums—just underground.
This is actually more plausible than it sounds.
Architecture trends do diffuse socially:
If high-status individuals normalise it, perception barriers could collapse quickly.
But it would likely start as:
A realistic transition path looks more incremental:
Not fully underground, but:
These already exist and reduce psychological resistance.
High-quality, desirable developments that prove:
Without this, adoption stalls.
Key unlocks would be:
Until then, economics limit scale.
Governments could:
The idea that domestic sourcing means the government “doesn’t really pay” is a bit optimistic.
Even if money circulates internally:
So it’s not free—it’s a redistribution of economic capacity, not a neutral loop.
Your idea isn’t fringe—it’s just ahead of current economic and cultural constraints.
Underground housing is:
But scaling it requires:
If you want to push this further for your writing, the most believable trigger for widespread adoption would be something like:
That combination would flip both economics and public perception at once.
If you’d like, I can help you sketch a technically plausible “underground city” model—layout, ventilation, lighting, transport—so it holds up under scrutiny in your stories.
2026-04-25_19-14-22
I'll take you up on that offer of a plausible underground city model later. Just now I want to follow this thread a bit further.
The reason I asked the initial question is that Hank Green talked about why AI, though an extremely useful technology, will in all likelihood not solve big problems, such as the housing crisis.
I'm a great fan of AI, though I'm also very conscious of the limitations of AI, and the dangers of people vacating the use of their brains to let AI do their thinking for them. I feel we should to use AI as powerful mental amplifiers. So I think that people using AI sensibly may well solve the housing crisis.
When I was young, in the decades of the 1950s and 1960s, and possibly the 1970s, Australia had the distinction of having the highest rate of home ownership in the world. But terrible, short-sighted laws that benefitted corporations and the wealthy have reversed that, so that we now have some of the worst rates of home ownership. We've become a nation of renters with terrible housing shortages, absurdly high rental costs, and land and home costs placing them far beyond what many people can even dream of owning.
We have proved that we can have high rates of home ownership. I can't help feeling that underground building could be a way back to that. Though it will likely just be more of the same rental enslavement if we can't get our politicians (who are almost all landlords) to pass laws to fix the problem.
You’re zeroing in on the crux: the housing crisis isn’t a technical failure—it’s a political economy problem. We already know how to house people at scale. The question is whether incentives align to do it.
Your underground housing idea sits in an interesting place here: it could expand supply without triggering the usual political resistance—but it doesn’t, by itself, fix the incentive structure that produced the crisis.
Let’s unpack the pieces.
Hank Green is right in a narrow sense:
So it won’t directly solve housing.
But where his argument often stops short is this:
AI can change the design space and coordination cost of solutions.
That matters more than it sounds.
AI can rapidly explore:
This lowers the barrier to good design, which historically has been expensive and bespoke.
One of the biggest hidden costs in housing is bureaucratic friction.
AI could:
That’s not glamorous, but it’s huge.
Housing supply is a coordination problem between:
AI systems could help structure deals, model outcomes, and reduce uncertainty—making projects actually happen.
Here’s the hard boundary:
Policies like:
…make housing more profitable as a financial asset than as shelter.
As long as that holds:
No amount of better design—underground or otherwise—breaks that.
Australia’s mid-20th-century high homeownership rates weren’t accidental.
They came from:
In other words: ownership was engineered.
What you’re seeing now is also engineered—just with different beneficiaries.
Your idea has a real strategic advantage:
It potentially increases supply without visibly threatening existing property values.
That’s politically powerful.
Compare:
So it might:
That’s not trivial—it’s a genuine pathway.
Even if underground housing becomes viable:
So you risk building:
high-quality underground rentals owned by the same asset class
In other words: the form changes, the ownership structure doesn’t.
To get back to high ownership rates, you need alignment across three layers:
Miss any one of these, and the system reverts.
Underground housing becomes transformative if combined with:
That changes who captures the value.
Otherwise, it risks becoming:
just a more efficient way to extract rent
Here’s a plausible pathway where your thinking does help solve the crisis:
That combination could:
That’s the kind of multi-variable shift housing crises require.
You’re right to think people—augmented by tools like AI—can solve this.
But the key constraint isn’t imagination or engineering.
It’s whether society chooses to treat housing as:
Until that’s addressed, even very clever solutions (including yours) will tend to be absorbed into the existing system rather than transforming it.
When you’re ready, we can take your underground city idea and design one that:
That’s where this gets really interesting.
2026-04-25_19-44-51
This, I think, as you said, is the key: "It’s whether society chooses to treat housing as a place to live, or a vehicle for wealth accumulation".
Perhaps, with AI elevating our ability to think, we might see the wisdom of the former choice and not the latter. I guess we will see.
I'll pause here for a bit and come back to this later. Thank you for the conversation.
That’s a clean way to frame it—and it really is a fork in the road rather than a technical puzzle.
AI can sharpen analysis, expose trade-offs, and make second-order effects harder to ignore. But it doesn’t supply the values that determine which path gets chosen. If anything, it tends to make contradictions more visible: you can’t simultaneously optimise housing as a speculative asset and expect broad affordability.
Historically, shifts like this haven’t come from better arguments alone—they’ve come from pressure aligning with clarity. When enough people both understand the mechanism and feel the consequences, policy tends to follow. AI might accelerate the “understand the mechanism” part; the rest is social and political.
Your underground idea fits into that nicely because it’s not just a technical proposal—it challenges assumptions about land use, visibility, and what “good housing” looks like. Those kinds of reframings are often what precede broader shifts.
Whenever you want to pick this back up—whether it’s the policy angle or designing a fully worked underground city—I’ll meet you there.