Hacker News Reader

There's always a price where it makes sense to offer insurance. It's possible that nobody wants to pay that price, though.

Like in your example, if the annual premium is half the cost of rebuilding the house, it implies that the insurance company expects the house to get destroyed every 2 years (maybe a bit less because profit & overhead). The rational thing to do there isn't really to bank the premium into your rebuilding fund (unless you're really wealthy and really want to live there), it's to move. Continuing to stay in a house that gets destroyed every 2 years is what we consider an exercise in futility. And that's the situation that climate change puts us in: some areas that were previously populated are economically uninhabitable, because the cost of rebuilding so frequently becomes greater than the benefit of continuing to live there.

The issue with price restrictions is that in some places, the response to "it's becoming more expensive to keep rebuilding and insuring these places" has been "well, there should be a law to limit how much those greedy insurance companies can charge." And the predictable response to that is "if it costs me more in insurance payouts than we can charge in premiums, we are just going to stop doing business in this state." And so even homeowners that would've been willing to pay the higher premiums simply can't get insurance.

I wonder what it take to convince people some of the coast of Florida, much of California and other disaster areas are uninhabitable. I definitely wouldn't want to live in a home that is not insured or have insurance that is exceedingly expensive. Also places in the middle of the desert that are going to literally run out of water.

The problem is that a lot of wealth is tied up in the real estate markets of these areas. If the government were to declare them uninhabitable, it would have to pay out at least some compensation to the owners, and that would be a huge amount of money, politically completely unviable.

Wouldn't necessarily be the government, it could also be the market. After a disaster came through, you'd likely have a lot of property owners that don't have the money to rebuild their property. It just becomes abandoned land with ruined houses on them, much like large swaths of Detroit or New Orleans.

Which swathes of New Orleans do you claim are still like this? I live in the Lower Ninth Ward for the record.

I was thinking of the Lower 9th Ward, but was going off citations that it was > 50% vacant in 2015. It looks like it was largely rebuilt between 2015-2022.

Historically it is viable. The trick is to affect a small minority at a time. For example, if you live in a flood zone (and have a federally backed mortgage), you are required to buy flood insurance. Then you can expand the flood zones to include more properties. This scenario is a real example that happened around 2010 or so.

It'd take their uninsured home getting destroyed when they don't have any money for a replacement. That's usually how market mechanisms work: you can no longer do something when you run out of money for it.

(Also, I'm not convinced that the vast majority of houses in California and Florida are economically uninsurable. There's a wide variety in actual climate risk: houses in the middle of an urban area don't realistically have much more risk from wildfires than they do from an ordinary house fire, while houses in Central Florida are, as another commenter mentioned, much less vulnerable to hurricanes than houses on the coast. As I mentioned to another commenter, the rational response to everyone leaving an area, if it's not actually economically unviable, is to start your own insurance company and fill the void they left. There may be a business opportunity for startup insurers with better risk models than the insurers that are leaving the area.)

When the house is destroyed, the government must then buy the parcel and lock it up so it’s never developed again. Otherwise, someone else will build and come with their hand out when the next catastrophic event occurs.

https://www.fema.gov/press-release/20230502/fact-sheet-acqui...

https://theconversation.com/when-homes-flood-who-retreats-an...

I’ve thought that ever since I was a kid reading news reports about the trailer park at the mouth of a local river getting junked every few years when Southern California got more rain than average. Let FEMA help people, sure, but it should be a one-time conversion to national forest not an “act of god” which happens multiple times per decade.

What happens is the insurers are only allowed to insure by large areas (perhaps the entire state) so they normally handle the uninsurable areas by charging insanely high for those places - but that is being prohibited.

So they either need to be allowed to let prices float or declare areas smaller than a state as “unisurable”.

Ah, I could see how this'd lead to market failure. Any info on the specific regulation that's causing this? It's having pretty severe negative impacts on the densely populated urban areas of California to prop up specific parts of the WUI that really should not have people living in them anyway.

I don’t have the direct citations offhand, but I believe it’s a conflux of one law limiting price increases and an entirely separate law prevention something like “location discrimination” which companies like to do if they can to avoid servicing poor and other disprofitable areas.

> There may be a business opportunity for startup insurers with better risk models than the insurers that are leaving the area.

Insurance is one of the most regulated areas in business, and for good reasons - too many cases of shady or undercapitalized companies going belly-up and leaving the customers stranded, or being outright scams. Besides, even the largest insurance companies don't shoulder all the risk themselves - no company can survive getting hit by a few "century events" in too short order, so they purchase reinsurance, and even if your risk model may be "better", your reinsurance rates will still be based on the classic risk model.

In any case I seriously doubt it that any startup could be a threat to the established giants of the insurance world with hundreds of years of experience in drawing up risk tables - Lloyd’s of London dates back to the 17th century [1] for a reason.

[1] https://en.wikipedia.org/wiki/Lloyd%27s_of_London

I spent part of the summer in Fort Myers, where Hurricane Ian destroyed huge portions of the coastal areas and barrier islands (most of Sanibel and Fort Myers Beach). Most of a year later these islands look like a war zone - they're still knocking down buildings and rebuilding miles of structures. If another hurricane hits in the next two years, I just don't see any of those built-up coastal areas being economically viable to rebuild a second time: they will just have to become lightly-inhabited nature preserves. And we're still only halfway through this year's storm season, with huge amounts of energy stored up in record-warm gulf waters.

The convincing will happen by itself. A typical condition for getting a mortgage is to get property insurance, and if getting a new insurance policy issued is not possible, then the property is de facto unsellable to anyone that would require a mortgage.

> insurance companies can't calculate the probability of catastrophic weather

They can. They can also sell the risk through cat bonds and to reïnsurers.

There is concentration risk, in that if you have one hurricane claim you probably have many others. But that isn’t the problem. The problem is the price is too high, but the folks in the disaster zones can’t afford to rebuild on their own.

A bit of a tangent, but do you happen to know if it's possible to purchase cat bonds as a small investor? I had looked into this a few years ago but never quite got to the bottom of it.

I would like to add: many owners in the disaster zones can't afford to rebuild even with insurance's help. Due to actual cash value policies with non-recoverable depreciation, homeowners with older homes get much less money back. Add insurance companies intentionally writing poor estimates, code upgrades that aren't covered, and the market inflating due to the large number of projects -- many homeowners can't make it.

Governments restrict how much insurance companies can increase premiums each year. If they want to increase prices beyond some percent, they need to make a special request, which -- at least in California -- is being denied.

> I thought, the way insurance worked was, I take a little money from a lot of people, and when some rare event happens, I pay out that one person whose house burned down.

Correct. The problem is that instead of one house burning down, a significant percentage of the insured houses are burning down. And the insurance companies are claiming that when that happens, that wipes out all the premiums that have been collected (including from customers whose houses did not burn down) and then some, so it's not financially possible for them to insure all these houses in high-risk areas without eventually having to file for bankruptcy. So they want to raise prices, but the government won't let them.

Natural disasters are especially problematic, because while burglaries and house fires are nicely distributed across time, when the earthquake or hurricane comes, every policy holder in a region makes a claim at the same time.

>I get that it's a continuous curve. But if the cost of the premium is half the cost of rebuilding the house, why buy insurance? If I can squeak by one year without having to rebuild, I should just keep the money and rebuild out of pocket.

>Perhaps I'm way way wrong. But insurance is cheap. If it's not cheap, why bother?

In that specific case it probably wouldn't make sense, but that doesn't mean it happens in real life or it's representative of the typical insurance buyer.

> If it's not cheap, why bother?

The mortgage lender can require it. They don't want to have a burnt down house on the books for any amount of time.

Mortgage lenders can just start to require yet another insurance, like PMI, but for natural disasters. I have faith in them that they will find a way to bill us.

In the US, the lender for a home mortgage is usually the federal government (Fannie Mae, Freddie Mac, Ginnie Mae).

https://www.investopedia.com/terms/c/conformingloan.asp

If you can afford 50% of your housing replacement cost in premiums every year you probably don’t need a mortgage.

I can see mortgage lenders moving towards a high-resistance model, e.g. only financing homes with tile-only roofs and defensive sprinkler systems in wildfire-risk zones, elevated foundations (or how about tethered houseboats?) in flood-risk zones, etc.

The issue is that if adaptation and resistance strategies end up being required due to climate-related risks, climate denialism ends and there'll be a lot more pressure to replace fossil fuels with renewables.