What do we know about the economics of storms?

That is the topic of my latest Bloomberg column, here is one excerpt from the final section:

Greater federal aid to the victims is a likely option. Still, it will not be obvious to many Americans why those who lost homes might end up receiving higher government benefits than people elsewhere who never had homes in the first place, and who may have inferior income prospects, especially if they live in rural areas. Many Texas congressional representatives voted against federal aid when Hurricane Sandy hit New York and New Jersey in 2012. That sounds cruel, but we also have to guard against the tendency for aid decisions to be driven by media cycles rather than the nation’s most serious structural problems.

One commonly discussed “economist’s solution” is to limit federal bailouts, and get the federal government out of providing flood insurance, in the hope that homes will not be built on risky floodplains. It doesn’t make sense to have $663,000 in payouts sent to one Mississippi home, which has flooded out 34 times in 32 years.

That all makes sense in theory, but it also illustrates the limitations of economics. The homes are already built in precarious places, and in the Houston area the damage has already been incurred. Are we prepared to forgo giving federal aid, just to set a precedent for the future? If anything, acting tough could lead to a backlash against the politicians who do so, and make it all the more clear that future bailouts will be forthcoming. Maybe the best we can do is to price flood insurance more appropriately, in recognition that climate risks are higher now. Over the longer run, as housing stocks turn over, that will help limit homeowner and also fiscal exposure to extreme events.

In other words, doing better in response to storms won’t be that easy.  At least so far, it seems the human logistical preparation and response has gone fairly well — relative to expectations — this time around.


Made some tweaking would move things from the current equilibrium to a better one.
1. Benefit lowered to 90% of the damage, not 100%. And to be lowered again if the premiums aren't covering losses.
2. Benefit caps. Say $100,000 per event and maybe another one for lifetime?
3. The deductible rises with each claim.
4. I guess the 90% could be lowered by 5% or 10% with each claim.
The point being to try to get people to take the money and abandon the houses in floodplains.
Oh, and put a sunset in for the federal plan so that we transfer this back to the private market.


Maybe it's politically unviable or even unfair to abandon coverage to existing homes. But why continue to subsidize new and rebuilt homes knowing what we're getting into. Furthermore we should cap coverage at present-day valuations. I.e. a home insured for $1 million is only eligible for that amount or less even if it's market value rises to $2 million. The government may protect people from losses, but they shouldn't be subsidizing their capital gains.

One final thing, this seems kind of like an issue like health insurance. It's not realistic to let uninsured people die from some medical emergency, and it may not be realistic to let the victims of big highly news-covered storms lose their homes. Realistically "society" should require insurance, because they're going to bail them out anyway. Any new home in a flood plain should either require market-priced insurance.

You could also add a sweetener to move. If you make a claim and want to put some of that money toward relocating, we'll up the reimbursement by X%. Or perhaps just cover moving expenses up to $5,000--enough to help. Once someone moves out, the premium increases for the next owner based on fact there's been a claim. Of course, for folks who stay, this risks turning those communities bad pretty quick as others clear out. But maybe that's the idea.

Aren't virtually all major cities built on floodplains?

More than should be, but no.

Is it not the case that 80% of the victims of Harvey did not have flood insurance? And yet their homes were built there anyway.

Why did the banks lend money for those homes?

What percent of the value of the homes are owned by mortgage holders?

Most likely the flood maps were outdated.

There is very little quantitative information about the flooding in Houston. Here is an example from Rice University:


Not a crisis, but many large nuisances.

There is lots of disaster porn in the news, but few summaries that explains the big picture quantitatively, for example what percentage of houses inside the loop have water intrusions, etc.

I think "inside the loop" is the good land that was originally developed as the city of Houston and is mostly in good shape. It's marginal land farther away that is not.

...good point -- most of these mortgaged/flooded homes are effectively owned by the banks and/or Fannie Mae, etc. The Feds are really "Bailing Out" the banks and Federal agencies, not the suffering residents.

Only federally-backed mortgages are required to have flood insurance in a floodplain.

Maybe the best we can do is to price flood insurance more appropriately, in recognition that climate risks are higher now.

Even that, of course, spawns backlash, as seen by the Biggert Waters Act almost immediately getting partially repealed ("gutted," some say), and that mostly affected second and vacation homes.

Being in Congress means making tough and tactical choices on the basis of how things are brought up to a vote (something that has always plagued legislators running for President, as one can ask Kerry.) When someone votes for aid in an early part of the process but then dislikes unrelated riders added to it, it does in a certain sense mean that they voted against aid, just as in the equal sort of criticism of John Kerry. While compromise should be encouraged and I understand the desire to attack deals, too much of this rhetorical trope only encourages the attempts to attach unrelated riders to everything. But perhaps people really want Planned Parenthood funding bans attached to every popular bill.

Note that gutting the Biggert–Waters Flood Insurance Reform Act of 2012 was done on a bipartisan basis, with almost all Democrats (despite supposedly believing in climate change) joining a majority of Republicans in voting to keep flood insurance from being priced more appropriately and delay reforms. So it's very hard to do the best we can do, and it invites deep backlash as well.

The backlash mostly comes from those with higher incomes, as the poor, as we've seen, don't even carry flood insurance-- but are probably right than in a large enough disaster, it will be hard to let them suffer too much.

It's as much a question of whom to bail out as it is a question of whether to bail out. If you refuse to bail out homeowners, you'll wind up bailing out the banks whose collateral was just destroyed. Switching to private insurance means bailing out the insurers, etc. Good luck finding a deep pocket who can shoulder losses of this magnitude without federal help.

Actuary science competently answers the questions of what-is-insurable, under what conditions and price.

"Government Insurance" ignores actuary calculation and makes its insurance decisions based upon politics. A financial mess always results eventually.

Government should not be in the insurance business anywhere !

This is a situation with some aspects of Talebs two main soapboxes, (1) skin in the game, and (2) tail risk.

A builder that makes a house that is prone to flooding, does not share the financial risk with the ultimate owner, and there are several indications that the flood maps are tainted by calculations based on assumptions of Gaussian distributions, and on the practical engineering level, flood maps might lag changes in developed lands that adversely affect water runoff.

Also, the decision makers in issuing building permits don't share the financial risk if the house flooded.

Having said this, there are indications that newer subdivisions of Houston did not have much problems with flood damage.

If taking a Taleb endorsed approach, what is the proper tradeoff between believing flood maps and use the minimum allowable elevation (100 years flood level plus 1 foot) or being more like Fat Tony and say: "Those civil engineers have no skin in the game, I will make the driveway 5 feet above 100 year flood stage, and elevate the house 10 feet above the driveway, using steel I-beams.

The advantage is a nice wraparound porch with a great view, and lots of cheap storage area under the house, or a nice shaded area for barbeque or growing some ferns. The disadvantage is that walking up a 10 foot stair might make the house unlivable for seniors or disabled people.

It is unclear whether public policy could or should force such an approach, but this approach makes lots of sense to me, and that is also how many homes in Galveston, TX are constructed.

"and elevate the house 10 feet above the driveway, using steel I-beams. ... might make the house unlivable for seniors or disabled people."

That will substantially add to the cost of the house. So it will likely make it unaffordable to anyone near the margin of being able to buy a house.

I realize there are costs, that's why I am not supporting an outright edict that all construction should be robust. However, with economics of scale, this style of construction doesn't have to be a big game changer.

I do think lots of US construction is simply a victim of long established bad habits. Why make wood framed houses in Florida and Texas? The standard (1970 through 2000) building style of single family homes in Norway was first floor is a concrete/cinder-block construction, second level is framed. If copying this style, and avoiding sheet rock, wood floors and wood paneling downstairs, the prep for a flood is carrying all furniture upstairs, and the cleanup is Clorox and a pressure wash.

Pure concrete houses would also withstand the wind damage associated with hurricanes.

"I realize there are costs, that’s why I am not supporting an outright edict that all construction should be robust. However, with economics of scale, this style of construction doesn’t have to be a big game changer."

Agreed, but anything that significantly raises the cost of the house will create push back. And Congress has a backbone of jelly.

Jelly? Too strong! I think low pressure helium!

In Australia all houses in the cyclone prone areas have to be cyclone proof, meaning steel frames, deep concrete footings, queenslander homes and be rated to withstand category five cyclones. Thats one reason why in northern australia hardly anyone lives there, but also why, when a cyclone hits, the damage in terms of homes and loss of life is very minimal.

Also, this "robust" construction will make the neighbors' houses less robust. (It happened to me, although the resultant flooding was not catastrophic.) It is hard to charge appropriately for this externality.

Partially agree. If I added material to elevate all of my property, then it would be marginally worse for the neighbors. If a whole neighborhood was elevated, then the same water amount would effectively occupy a smaller area, affecting nearby neighborhoods. This is the same problem with levies all over the flat river plains of the Mississippi and Missouri rivers.

I can see that if your neighbor elevated his land to remedy flooding, and the drainage went straight onto your property, it would affect you. That wouldn't be very neighborly.

Are you in Houston? I lived there 4 years in the nineties.

Are you saying beefy mustaches are not hearty enough.

I have heard that elevators are down to $12,000 to install. https://www.bing.com/search?q=home+elevators+prices&qs=LS&pq=home+elevators&sk=LS1&sc=8-14&cvid=7A504DF8F2EE452FBEA2C093EDBD1CF0&FORM=QBRE&sp=2

I have a home in a hurricane prone area, so I have experience with flood insurance. For a brief period following enactment of Biggert-Waters, as owners anticipated the end to federal subsidies for flood insurance, two things happened. One, homeowners in flood zones panicked, fearing a spike in flood insurance premiums (rather than several thousand dollars a year, premiums would explode to $10,000 to $20,000 a year for what is a modest home), resulting in a spike in supply of those homes in the market. Two, demand for those homes collapsed, as potential buyers feared the same rise in flood insurance premiums. One could argue, and many do, that it's too bad: those who purchased homes in flood prone areas should be forced to pay for their folly. That may be fine in theory, but there are millions, tens of millions, of people who committed the same folly. A sharp rise in flood insurance premiums would make many homes unsaleable, precluding their owners from moving (for a better job, etc.), causing a plunge in their savings (as reflected in their homes), and causing a spike in mortgage defaults and bankruptcies (because the homeowners couldn't afford the flood insurance and couldn't sell their homes). I've commented before that I would not buy a home in a hurricane prone area today. Yet, developers and builders in my area can't construct houses fast enough to satisfy the demand. Are these people nuts! And it's not just the potential for sky-high flood insurance premiums but the devastation from a hurricane. Indeed, a hurricane passed through the area about a year ago, causing wide-spread damage and enormous inconvenience (for those lucky enough not to have suffered damage). Will scenes from Houston cause people to re-think buying a home in a hurricane prone area? For some, but not for the vast majority who fall prey to the lure of life near the ocean.

Is it possible to make a distinction between homes that have been there a long time and newly constructed buildings.

I'm vey dubious about subsidized flood insurance, but it's even worse when it is provided for new construction.

If the premium is not negotiable, then one can return to equilibrium as, in the event of a flood, X% of the loss reimbursement becomes a non-forecloseable lien against the property.

They could lift the home, that should drop the insurance premium. That is you take the cheapest route rather than just pay the premium. Much of the value is in the land.

People panic too soon.


I am in sympathy with you on this, rayward, as I prefer a wild coast, but feel compelled to point out that the people whose livelihoods involve stuff like this ...


... are not living on the Gulf for "island time" and margaritas.

Oil and gas is under the aegis of the state Railroad Commission (don't ask), and the chairwoman says supplies are plentiful. Her credential is she's the daughter of a longtime statehouse Speaker. [If I recall the Lege wrote a law just for her, one time, allowing the children of legislators to keep their generous state health insurance plans into their thirties.] So that's fine. But the pumps are completely dry in Austin and Dallas; you might want to go fill up your tank just as a wholly unnecessary precaution.

They have restarted the refineries, though.

Indeed. The plants (like Arkema's) are close to the refineries (for their feedstocks or to feed catalysts, etc. to other chemical plants that draw base chems off the refineries) and the refineries are close to the ports (for the obvious reason). And all of it is basically in a giant marsh. People tend not to know that malaria was endemic in the area before WWII but if you do and if you're ever been on say Pleasure Island in Port Arthur when the Sun goes down and the mosquitos come out you'll realize that the value of the land, sans industry or oil/gas production, is basically zero. For those of you not from around here the song "Anahuac" by the Austin Lounge Lizards will give you some idea of what life is like for those who live in the SE Texas and southern Louisiana coastal marshes: https://www.youtube.com/watch?v=A1GoDtYqjT4

As I commented to a different blog post, recurring payouts for the same home are because the living area of the home is below the flood plain. These typically are older homes, built before local rules required the living area to be above the flood plain. Local rules can mitigate this phenomenon (local rules typically require a home be brought up to current code, including elevation, if more than 50% of the home is damaged), but enforcement is lax (not least because local lenders, appraisers, etc. help "avoid" the 50% rule). What the national flood insurance program could do is adopt rules that make local enforcement more likely and use the 500-year flood rather than the 100-year flood as the point of reference for the "flood zone". Many politicians would object, arguing that building codes are a matter of local concern not federal. Yea, right. .

We should know though that if we dig 62.5 leverages deep channels to what’s AAA rated, and 12.5 deep ones to “risky” SMEs, even though both live in the same real economy, then the former will sooner or later suffer inundations, and the latter will see drought


Herod feathers are plume-like hydroids,
Rillilking in the Basquiat of harvesty.
Pioneers! Cherubs! Scions! Such rarefied rays.
Ereing ions.
Tussut beiru
such sneer and cheer and beers and deers, seers and peers,
skerries and pixies, the faint dust an echo
Of home ruts, stirred pots.
Relish it, he rode the kill.
Of the muts and must stains,
the sumptuous armor of mercury,
And Mildew.
The dainty stuff of oohs and aahs.

Also, if one uninsured house floods the home owner and mortgage holder pays, if hundreds flood the taxpayer pays.

"in recognition that climate risks are higher now": pathetic drivel. You should be ashamed of yourself.

I kind of liked that part. "Do you really believe that climate risks are increasing? Then pay up."

The Texas Seed Bill:

President Grover Cleveland once told a friend that he saw his chief legislative duty to be stopping bad bills from becoming law, rather than trying to convince Congress to pass what he thought would be good ones. Cleveland used the veto more than any other US president before or since.

One of Cleveland’s most famous vetoes was his veto of the Texas Seed Bill in 1887. A long and severe drought had stricken areas of Texas. With no grass to graze, eighty-five percent of cattle in the western part of the state died. Those cattle that remained were starving, often motherless calves. Many farmers were also close to starvation and had eaten their seed corn to survive. Congress authorized a special appropriation to send seeds to the drought-stricken farmers. The amount ($10,000, or approximately $223,000 in today’s dollars) was small and the need was great, but Cleveland vetoed the bill.

His veto message expressed his commitment to the Constitution and the importance of private charity. He said that while he thought the intentions of the bill were good, he had to withhold his approval. He wrote,

“I can find no warrant for such an appropriation in the Constitution, and I do not believe that the power and duty of the general government ought to be extended to the relief of individual suffering which is in no manner properly related to the public service or benefit. A prevalent tendency to disregard the limited mission of this power and duty should, I think, be steadfastly resisted, to the end that the lesson should be constantly enforced that, though the people support the government, the government should not support the people.”

Furthermore, Cleveland said, it would weaken the “bonds of a common brotherhood” for the government to provide assistance to individuals where individuals, families, communities and private charities otherwise would.

If we had more leaders like this the world would be a far better place.

"One commonly discussed “economist’s solution” is to limit federal bailouts, and get the federal government out of providing flood insurance, in the hope that homes will not be built on risky floodplains"So that's it: the richest country in the world will just forsake its own citizens. Let them drink flood! The bank of justice is bankrupt, there are insufficient funds in the great vaults of opportunity of that nation. It is hard to believe that.

...the FEDs are bankrupt now (100 TRILLION+ debts) -- they have nothing to pay their own internal expenses except a printing press. So nothing left for Texas but funny-money.

What percent of your personal assets have you now donated to the Texas needy?

So that is it: there are billions of dollars for bombing weddings and spying on the populace, but there is no money for Americans citizens facing an act of God. Again the Roman story."The wild beasts of Italy have their caves to retire to, but the brave men who spill their blood in her cause have nothing left but air and light. Without houses, without settled habitations, they wander from place to place with their wives and children; and their generals do but mock them when, at the head of their armies, they exhort their men to fight for their sepulchers and the gods of their hearths, for among such numbers perhaps there is not one Roman who has an altar that has belonged to his ancestors or a sepulcher in which their ashes rest. The private soldiers fight and die to advance the wealth and luxury of the great, and they are called masters of the world without having a sod to call their own." American workers have been betrayed by their leaders.

If you want to get rid of federal flood insurance, I want to get rid of federal crop insurance. You gore my cow, I gore yours. Where does it end? How about getting rid of the FDIC? Instead of thinking of terminating popular programs protected by powerful politicians, you should be thinking about creating new programs. A dramatic fall in the stock market can be devastating to the retirement investments of many people -- why don't we have a federal stock insurance program?

You hit the nail on the head.

Also, these programs are the gifts that keep giving, from the taxpayers pockets:



The liability is not finished with the disbursement, it just keeps going.

White House Homeland Security Adviser Tom Bossert said at the White House press briefing that we’ll see 100,000 homes affected by storm Harvey.

Isn't this what the terrorists have in mind? Are Al Jezeera watchers toasting one another with cola drinks? Like Florida college professors are Islamic fanatics observing that this is a spiritual punishment for Texans casting their ballots for Trump? Maybe some Predator sorties could put the lights out on their parties.

Bloomberg: "Ty, I'm going to need a piece that implies that Trump is ripping taxpayers off by having the Feds do exactly what they've always done -- the very same thing we celebrated in the Obama years. Federal relief for disasters is now PROBLEMATIC. And here's the critical part: I need you to slam Texas politicians at the same time. Can you do it?"

Cowen: "On it, sir!!"

In which OJ cries about the partisan crap he and his posse never ever do when a Dem is president. He's not a hypocrite, he posts just as many things ragging on Trump's use of a teleprompter and taking vacations as he did when Obama was prez.

It should be pretty simple. If the insurance pays a claim, the fed seizes the underlying land and nothing can be built there again for the next thousand years. The prior owners can buy a house somewhere else.

If you want to be less strict it could be built, but the parcel is barred from insurance and can only be built on by ''rich' entities who deposit 10x-100x the cost of the structure in escrow so there is no temptation to bail them out.

And how do exports and imports move out and in absent coastal communities occupied by, you know, people. Of course, the people necessary to move exports and imports attract other people, including poor people who service the needs of the not poor. We live in an era in which ignorance is confused with intelligence, but really, does it have to be at this blog? The people living on the front lines of exports and imports are heroes, accepting the risks so that the rest of us have more enriched lives. Maybe if Cowen had lived in that dorm he would be more empathetic to others and less obsessed with what's most efficient.

There aren't zoning laws in Houston so anyone can build anywhere and this includes the areas originally set aside as drainage fields. So much for the libertarian paradise!

I don't know if this has been thrown out there, but rather than a payout, offer a buyout. If the owner pays the premium, the benefit is that the government buys your property. Once it owns your former property, it is required by law to demolish the house, cap off the utilities and convert the space into a park or water retention. Difficult to implement I think, but is it a bad idea??

Lehmann Brothers was abandoned, left to sink or swim--heavily weighted toward sink--pour encourager les autres (comme on dit). How well does setting an example work in practice?

Get the gubmint outta da flood "insurance" bidness and let the free market work it's magic. Right now, gov ins is a moral hazard. Maybe people will move away or build to handle flooding cheaply.

I keep hearing about that house in Mississippi.

I think that it has become the "Welfare Queen" of flood insurance, where a single memorable case of abuse becomes the argument to shut it down for everyone.

And who says that paying high costs for insurance isn't a net win? Before this storm this and other blogs were full of complaints about building restrictions driving up the price of housing.

What if the choice is periodic very high insurance costs, or constant unavoidable homelessness and economic malaise caused by people being unable to afford dwelling near their work and other resources?

You cannot usefully say "we need safety and insurance restrictions!" and also "we need plentiful cheap housing near employment, medical, and education centers" at the same time!

My simple nudge would be to require signs posted every 100 yards along any street in the 100 year flood plain saying that the area is subject to frequent floods.

Suppose we mandated all homeowners policies to cover flooding. But....

If you have a flooding claim worth more than 15% of the home's value, that is strike 1. If another claim appears in 5 years or less, that is strike 2 and the insurance company can, if they choose, double the premium. If another claim appears in 5 years after that they can then cancel flood coverage entirely and revert you to non-flood coverage.

Premiums would be affordable and market based because the cost of flooding will be spread out among many homes that never get flooding. If you get a rare storm that swamps a huge number of homes, you're not going to have a problem with people suddenly being locked out of coverage.

But if you build or buy a house in an area that gets frequent flooding you're not going to last very long with coverage. Perhaps specialty insurance will develop where companies are willing to cover dangerous houses based on customized inspections and evaluations.

1. You don't have to worry about revising flood maps or accounting for climate change. As areas become more flood prone, the market will adjust and treat them as such as claims start becoming more and more frequent.
2. No need to subsidize flood insurance at below market rates. The market sets the price but insurance companies can't use a 'pre-existing condition' of being in a suspected flood plain against the homeowner but that changes as real claims are made.

Comments for this post are closed