Wouldn't installing canopy systems be a more cost efficient solution to this problem?
Let's do some math, The dealership listed in the article (Rickenbaugh Automotive Group) currently has ~536 cars listed on their website(s) for sale. I'm going to assume that this number is a little lower than what is actually in inventory given that there are trade-ins and they probably don't restock inventory during the middle of the month. So I will round up to 600 cars to give them wiggle room for expansion and sales. Given their insurance policy is costing them $600k per year we are approximately spending $1,000 per car per year for insurance.
There seem to be companies that make canopies for this exact situation. While I can't find pricing I would have to imagine that the ROI looks pretty attractive given the sheer cost of their insurance and the simplicity of the materials involved in building a canopy. (metal beams and mesh netting)
Why not just solve the problem entirely instead of coming up with an exotic insurance policy?
It's a good idea to solve the problem instead of what the article describes, but there's an issue with canopies.
The types of storms that produce hail, usually supercellular, often have damaging straight-line winds. While out storm chasing, I've seen several gas station canopies collapse from strong winds - I'm struggling to think of a canopy solution that wouldn't have the risk of collapse.
Then you end up insuring your canopies against wind damage... And your cars against the possibility of canopies falling on them. And still insuring your cars against hail in the event that the canopy blows off, or the hail comes down at an angle and isn't fully blocked by the canopy, or whatever.
Also, I'm imagining canopies getting in the way of 1) making inventory highly visible (both from the road and other parts of the lot) and 2) making it harder to pack cars really close together because you have to deal with all the canopy supports.
> I'm struggling to think of a canopy solution that wouldn't have the risk of collapse.
A canopy that isn't solid and allows the wind to pass through. A mesh or screen should do the job. The opening size can also be larger so perhaps even netting with openings upward of a few cm could work.
I've also seen a lot of bad construction that catches on wind like a sail. It doesn't have to be built this way, however. A canopy that protects from hail coming from above can still have lots and lots of holes that lets wind go through.
Wind driven hail traveling nearly sideways is common unfortunately. I'm not sure what angle of protection would be the best compromise, but I do know a copy above would not afford 100% protection.
Hmm, I don't think I have experience with such hail. In my experience, the bigger they are and the harder they hit, the more vertical their trajectory. But my area isn't particularly prone to hailstorms.
Allow me to nerd out a little, since my other love besides development is severe weather.
The forward flank downdraft or rear inflow jet of a linear convective system/bow echo can see winds of 50mph+ rather commonly. The more interesting part IMO, is the rear flank downdraft, because it is closely related to tornadogenesis and can also have winds up to 100mph. The RFD is co-located with the hail core in a supercell, so some hail will get caught up in this wind that exists even when there isn't a tornado.
Now terminal velocity of your average baseball-sized hail is roughly 100mph, but you can imagine with 100mph horizontal winds that there can be a strong non-vertical component to the hail in certain portions of the storm. I've personally been hit by smaller 1-2" hail that is traveling nearly horizontal, and I've seen evidence of larger hail impacting at the same angle.
It's a safe assumption that the larger the hail is the more damaging it is, but not all hail is created equal! Depending on how accretion happens and the warmth of the atmosphere, you can have very hard 1" hail that damages cars, and very soft 2.5" hail that doesn't damage cars. My personal rule of thumb is that you need to be wary of anything 1.75"/golf ball or larger.
Having lived in tornado alley, I've seen softball sized hail completely punch through a residential roof (shingles, and plywood) and then punch through the sheetrock in the ceiling.
Maybe a soft one-inch foam canopy secured closely to the vehicle so wind can't catch it, with couple straps under the vehicle chassis? Could that work? Could it be put in place in time?
Alright, that's not really a canopy, more like a shield or body armor.
I'm not sure you realize the amount of kinetic energy a piece of ice the size of a baseball has at terminal velocity. Short of some sort of thick steel box.
It wouldn't be hard to make some thick foam mats with straps sewn to them that can be rolled out and fastened to cars in under a minute each. Maybe toss in a couple plastic strips to make it unroll itself like the reverse of a slap koozie. Weather forecasting is pretty good. It's not like you don't have decent warning. You can cover everything out back and leave the cars that are on display out front uncovered until the last minute.
Tossing a mat over a high roof van or large pickup wouldn't be easy but the hordes of crossovers could be well protected.
Alright! StormShield, YC S20, is officially launched! First round starts at $20m!
OK, maybe it does almost sound feasible, but I wonder what someone with outdoor car dealership experience would say? Is this a practical idea? How often do you need to unroll those mats?
This would also help with snow in the winter! I have to imagine it is a lot of work clearing snow off of cars for test drives, if you could just roll most of it off that would help!
I had this idea. I had a new car in 2014 and I had to park it outdoors for awhile while I was working on a different car in my garage. Anyhow, I didn't want hail damage so I put a car cover over the car, then piled a bunch of cardboard on top, and then threw another car cover over that. It worked until it got really wet, then I had a ton of soggy cardboard to deal with. But it worked as far as protecting the car from hail goes.
I was musing at the time why someone on Amazon wasn't selling some kind of hail protection padded car cover.
I have tons of great ideas, but I never connect on them. Then I watch other people make a fortune off that same idea. Shrug...
Exactly. This "God's Joystick" essentially just gives the insurance industry more data to build more accurate risk models. It's essentially pricing the market more perfectly without solving the actual problem itself which is hail.
The article should have pointed out to the reader that this should bring premiums down in areas that have so little chance of hail that it is not worth building physical protection. Right now, the data is not granular enough so insurance companies spread the cost over areas that they shouldn't.
I think a thicker padded car cover would be better, something akin to grill covers with a half inch padded lining. You can have them color coded for the type of vehicle. Grey for sedan, black for pickup truck, dark blue for minivan/hatchbacks. When a storm is coming they can just roll them out in a shopping cart and throw them on car by car. You can start a day or two before, covering about 80% of the cars (the ones that are less popular) so you can still be open and selling, and then the day of/before you cover the rest.
Let's apply some old school tech. Forget about insurance and canopies, think Atari Missile Defense.
A swarm of small pulse lasers surrounding lots, along poles on the inside, and generally build a defense network that takes out hail on the way down. Design it after Israel's Iron Dome. Sure, the targets are smaller, in a adversarial environment, and don't have heat signatures but it would be amazing. :)
Gut reaction: Because Ice/Snow needs an incredible amount of energy to go from solid to liquid, the lasers would have to be melt-flesh-in-an-instant powerful. Not so good for the eyes of airplane pilots.
I live in a city with big, annual summer hail stones. Many dealerships have built canopies and it seems to be working.
I also have friends in the auto industry and the interesting thing is they said hail isn't the worst thing, they actually profit quite a bit. Basically insurance will appraise the new value of each car and write a cheque (ex. $10k for each car). The dealership can still sell the car as "hail damaged" and offer the buyer a $5k discount. Buyer thinks they are getting a great deal, the car just has some dimples, but $5k is $5k right. The deal then makes an additional $5k off the car. However this breaks down if hail insurance premiums are going up.
Dealers can take the money, fix the damage and still sell the car as brand new - if it hasn’t been titled yet, body work doesn’t need to be disclosed to the buyer. Dealers usually have in house body shops or relationships with body shops and pay labor + supplies in the repairs. A bigger issue is the backlog of work that quickly accumulates in the aftermath of a hailstorm.
I imagine "the market" would correct for that profit. The $10k for each car comes from their insurance premiums. If they don't need the full payout, it make sense to want to pay less in premiums.
>"Two years ago he paid about $160,000 to insure $20 million worth of Cadillacs and Volvos on his lot, where about a third of the inventory is kept outside. This year he’s paying almost $600,000."
Surely a dealer could install some hail-proof canopies for a lot less than $600,000?
A little weird I can answer this, but I bought my Volvo through this exact dealership.
They have a very space constrained layout especially for inventory display, as their dealership is at the junction of some odd angled non grid streets (namely Speer).
I.e they’re an urban dealership that just doesn’t the have room.
Am I missing something or will this make no real difference?
Ultimately if there is a hailstorm then the insurer has to pay out, premiums have to cover that. Measuring whether there was a hailstorm or not might help to prevent fraud. Fixed payouts would help to reduce the cost of accessing damage. But neither change the core maths here: more, heavier hail means more payouts, means higher premiums.
Yeah, reading the headline I was suspecting it would be something that would erect a cheap weather-proof roof or something like that. Why not put on inflatable car covers when there's bad weather coming up?
Or invest in a bigger showroom or a roof without windows; if the insurance premium for not having that is 100's of K's a year then it makes financial sense to invest in something like that.
Can anyone help me understand how having this data actually lowers insurance premiums? The same number of cars are being damaged by the same number of hailstorms with or without them.
You want to look up adverse selection and moral hazard in an economics text about insurance.
Basically if you have some mixed population where some are lower risk and some higher, and they're separable, the low risk people will decide the insurance is not worth the premium. That leaves the higher risk population on the books, which will cause problems for the insurer of they had assumed they'd get a nice sample of the population.
So maybe some car dealers are aware of possible hail storms in their region but for some reason their particular shop is protected. Maybe by hills that drop the hail on the other side or something that which gross aggregates might miss.
Moral hazard plays in as well. If you're protected maybe you place more cars outdoors than you would otherwise, raising the costs to the insurer.
I can see where relying on a black box could get really hairy though. What if it’s a light storm that produces virtually no damage but the device produces a payout? What if the exact opposite happened? Who is programming and vetting the code?
Great points. Basing a damage payout on anything besides actual damage feels counter-intuitive.
Insurance companies already know where heavy hail falls - there are several companies with products that correlate hail reports with the dual-pol radar Hydrometeor Classification Algorithm product.
Any sort of black box won't be able to assess the hardness of hail, the distribution of hail or how the wind is driving it. I've seen damage from dense, hard 1" hail cores, and I've driven through softball hail thrown out the back of a storm without any damage.
ok, so a canopy is likely to catch the wind and possibly cause additional damage. But what about a net canopy system. Say, perhaps with 1cm sized holes. Big enough holes that the wind can't do a ton of damage, but small enough that the really disastrous hail gets caught and the mid size stuff gets slowed down as it likely deflects off the net. Anything that gets through undeflected is unlikely to damage the vehicle.
Money is cheap to borrow and selling used cars is fairly profitable. Better to sit on a car for an extra 6 months, paying 2-3% than lose a sale entirely.
Drones fly around and use ML to recognise hail, and when it detects hail, sends a signal to self-driving cars to (cue Monty Python's Holy Grail:) "Run away! Run away!"
Let's do some math, The dealership listed in the article (Rickenbaugh Automotive Group) currently has ~536 cars listed on their website(s) for sale. I'm going to assume that this number is a little lower than what is actually in inventory given that there are trade-ins and they probably don't restock inventory during the middle of the month. So I will round up to 600 cars to give them wiggle room for expansion and sales. Given their insurance policy is costing them $600k per year we are approximately spending $1,000 per car per year for insurance.
There seem to be companies that make canopies for this exact situation. While I can't find pricing I would have to imagine that the ROI looks pretty attractive given the sheer cost of their insurance and the simplicity of the materials involved in building a canopy. (metal beams and mesh netting)
Why not just solve the problem entirely instead of coming up with an exotic insurance policy?