Is the world becoming uninsurable?

Not uninsurable, but buildings are going to have to become tougher.

It's happened before. Chicago's reaction to the Great Fire was simple - no more building wooden houses. Chicago went all brick. Still is, mostly.

The trouble is, brick isn't earthquake resistant. Not without steel reinforcement.

I live in a house built of cinder block filled with concrete reinforced with steel. A commercial builder built this as his personal residence in 1950. The walls look like a commercial building. The outside is just painted cinder block. Works fine, survived the 1989 earthquake without damage, low maintenance. It's not what most people want today in the US.

If the market is allowed to price insurance correctly then we can motivate building designs to be more disaster resist. If the McMansion can't get insurance but disaster resistant, modest homes do, then people will adapt.

"Correctly" is doing a lot of work here. Some readers might miss that this is double edged. Insurance is a mandated product. You don't have a choice if you want a mortgage, or want to run a business. So while it is true that the sustainable price for insurance in many areas is higher than what current regulations allow, let's not forget what happens in an unregulated insurance market; price gouging.

If the regulators have defined 'price gouging' as a price substantially below the break even mark, literally any profitable insurance product is implicitly believed by them to be price gouging. The US does a weird thing where "insurance" no longer means pooling risk but some sort of transfer payment welfare system. If they're going to define "price gouging" as profitable activity it is hard to see how the economy is going to function.

Allowing insurers to make a profit and run a business without interference is going to be cheaper - and in most instances better - than whatever the politicians are trying to build here. If you get rid of all the mandatory-this and price-gouging-thats then to stay in business insurers have sell products that people want to buy at a competitive yet sustainable price. It works for food, it'd work here too.

The math of insurance suggests that, if it needs to be widely carried (either due to things like mortgage requirements, or the simple realization most people don't have enough resources to absorb a major catastrophe themselves), the most economical way to go is to have a single risk pool that's as broad and diverse as possible, so it can swallow a large clustered crisis more easily. Yes, this is a bit of robbing Peter to pay Paul.

I always found it funny when insurance marketing talks about "personalized rates", when the goal is to DE-PERSONALIZE the risk. If you have 10,000 customers in Los Angeles, and 5 million elsewhere, you can either isolate the LA customers and charge them the "real" price of the risk, which will be unviable as a business and probably politically touchy too, or you can include them in the broad pool, and the people with a full-cinderblock home in a non-flammable state pay $20 more a year so the entire endeavour can work.

The concept probably works better if you have some concept of social cohesion to lean on-- you might not get the best possible outcome personally, but the system itself is more robust for everyone.

This completely ignores incentives. If insurance isn't allowed to charge people more who live in fireprone or floodprone areas, more people will live in such areas, and overall society will have to spend more money rebuilding when disasters inevitably hit those areas. Personalised insurance pricing would allow insurers to charge much more to people living in such areas, which incentivises people not to live there. It's also a moral issue: if everyone pays the same rate, then people who did the right thing and chose to live in an area that wasn't fire or flood prone are subsidising people who did the risky thing.