When to self insure for things like cars? (prefer mathematical model)

 Posts: 76
 Joined: Tue Oct 18, 2016 1:12 pm
When to self insure for things like cars? (prefer mathematical model)
Let me give you an example. Let's say you had 10k to your name and your had a 15k car, buying comprehensive and collision insurance should be a no brainer. Now let's say you had 100 million, obviously you wouldn't need insurance on a 15k car, you'd just write a check without blinking. What is the threshold on when you should go without insurance?
My gut instinct is that if your net worth is 10x the value of what you're insuring, you should self insure, but it's not really based on anything. Does anyone know of any mathematical models of when to self insure for items based on your current net worth?
My gut instinct is that if your net worth is 10x the value of what you're insuring, you should self insure, but it's not really based on anything. Does anyone know of any mathematical models of when to self insure for items based on your current net worth?
Re: When to self insure for things like cars? (prefer mathematical model)
Liability!!!adherenceEnergy wrote: ↑Wed Jul 22, 2020 7:37 pm Let me give you an example. Let's say you had 10k to your name and your had a 15k car, buying comprehensive and collision insurance should be a no brainer. Now let's say you had 100 million, obviously you wouldn't need insurance on a 15k car, you'd just write a check without blinking. What is the threshold on when you should go without insurance?
My gut instinct is that if your net worth is 10x the value of what you're insuring, you should self insure, but it's not really based on anything. Does anyone know of any mathematical models of when to self insure for items based on your current net worth?

 Posts: 76
 Joined: Tue Oct 18, 2016 1:12 pm
Re: When to self insure for things like cars? (prefer mathematical model)
It depends if your collision/comp includes window glass with zero deductible.adherenceEnergy wrote: ↑Wed Jul 22, 2020 7:45 pmFor sure. I personally get max liability insurance and umbrella insurance on top of that, but don't carry comprehensive or collision. I'm just curious if there's any rule of thumbs or math involved in when to drop comprehensive and collision.
Re: When to self insure for things like cars? (prefer mathematical model)
I assume the discussion is about self insuring your own vehicle (dropping collision), not the liability.
If you have a 15K car and can afford to replace it, and if you are a reasonably safe driver, in theory you should self insure and not carry collision.
In real life though, things are different. I don't self insure but I have a large deductible  2500$. I found from experience that if the damage to my car is under 2500$, my insurance company has no interest in doing anything even if I had no fault in the accident. The other insurance (if exists) might be reluctant to pay. What do you do then? Hire a lawyer? Waste your time in small claims court?
If you have a 15K car and can afford to replace it, and if you are a reasonably safe driver, in theory you should self insure and not carry collision.
In real life though, things are different. I don't self insure but I have a large deductible  2500$. I found from experience that if the damage to my car is under 2500$, my insurance company has no interest in doing anything even if I had no fault in the accident. The other insurance (if exists) might be reluctant to pay. What do you do then? Hire a lawyer? Waste your time in small claims court?

 Posts: 2008
 Joined: Thu Jul 09, 2015 7:00 pm
Re: When to self insure for things like cars? (prefer mathematical model)
+1.Starfish wrote: ↑Wed Jul 22, 2020 7:50 pm I assume the discussion is about self insuring your own vehicle (dropping collision), not the liability.
If you have a 15K car and can afford to replace it, and if you are a reasonably safe driver, in theory you should self insure and not carry collision.
In real life though, things are different. I don't self insure but I have a large deductible  2500$. I found from experience that if the damage to my car is under 2500$, my insurance company has no interest in doing anything even if I had no fault in the accident. The other insurance (if exists) might be reluctant to pay. What do you do then? Hire a lawyer? Waste your time in small claims court?
It also depends on how much the collision insurance costs. If you’re elderly with a flawless driving record it might be pretty cheap.
I would say asset < 50x net worth is closer to a selfinsure rule of thumb than 10x.
The most precious gift we can offer anyone is our attention.  Thich Nhat Hanh

 Posts: 7106
 Joined: Thu Oct 15, 2015 3:53 pm
Re: When to self insure for things like cars? (prefer mathematical model)
Everyone has different levels of risk tolerance, so just like an AA, there is no formula. As others pointed out, your liability is almost unlimited regardless of what the house/car is worth. Unless you have tens of millions, I wouldn't consider self insuring (normal thingsnot counting getting collision on a $3k car).

 Posts: 5047
 Joined: Wed Apr 08, 2015 11:31 am
 Location: West coast of Florida, inland on high ground!
Re: When to self insure for things like cars? (prefer mathematical model)
If you are driving a $15,000 car with $10,000 to your name, you cannot afford to self insure.
And, that goes double if you are still toting the note.
Being retired, having only liability, and having to spend money to repair or buy another car wouldn't break me.i have more than $10,000.
If you are financing a car you might be under water, paying the balance of the loan if the insurance check doesn't pay enough. And you still need another car.
Maybe a boglehead who sells car insurance could educate us so far as what percentage the liability portion represents in the premium.
Broken .an 1999
And, that goes double if you are still toting the note.
Being retired, having only liability, and having to spend money to repair or buy another car wouldn't break me.i have more than $10,000.
If you are financing a car you might be under water, paying the balance of the loan if the insurance check doesn't pay enough. And you still need another car.
Maybe a boglehead who sells car insurance could educate us so far as what percentage the liability portion represents in the premium.
Broken .an 1999
“If I cannot drink Bourbon and smoke cigars in Heaven then I shall not go. " Mark Twain

 Posts: 1488
 Joined: Tue Mar 05, 2019 10:29 pm
 Location: Colorado
Re: When to self insure for things like cars? (prefer mathematical model)
I carry max liability insurance but no comprehensive or collision. My car is worth about $6000. If it was totaled tomorrow, I have more than enough cash to just buy another one (net worth in the mid six figures).
 Artful Dodger
 Posts: 1053
 Joined: Thu Oct 20, 2016 12:56 pm
Re: When to self insure for things like cars? (prefer mathematical model)
A few years ago I was thinking about dropping the collision coverage on a car my daughter drove. I never got around to it, and a few months later she totaled it. Truthfully, the insurance company valued the car over what I thought it was worth. The company paid me $4500 ($5000  the $500 deductible). I thought the car was worth maybe $3500$4000. I'm super happy, I didn't drop it.
I tell you this to give you a window into my thought at the time. At the renewal, the cost to have the collision / comp was about $380 a year above the basic liability policy. I didn't mind paying 1/10, maybe 1/12 of the expected return for the coverage, in this case $380 for the $4000 benefit was in that range. As we were crossing that threshold, I was considering dropping the coverage.
I tell you this to give you a window into my thought at the time. At the renewal, the cost to have the collision / comp was about $380 a year above the basic liability policy. I didn't mind paying 1/10, maybe 1/12 of the expected return for the coverage, in this case $380 for the $4000 benefit was in that range. As we were crossing that threshold, I was considering dropping the coverage.
Re: When to self insure for things like cars? (prefer mathematical model)
"Self insurance" is an oxymoron.adherenceEnergy wrote: ↑Wed Jul 22, 2020 7:37 pm Let me give you an example. Let's say you had 10k to your name and your had a 15k car, buying comprehensive and collision insurance should be a no brainer. Now let's say you had 100 million, obviously you wouldn't need insurance on a 15k car, you'd just write a check without blinking. What is the threshold on when you should go without insurance?
My gut instinct is that if your net worth is 10x the value of what you're insuring, you should self insure, but it's not really based on anything. Does anyone know of any mathematical models of when to self insure for items based on your current net worth?
 lthenderson
 Posts: 5221
 Joined: Tue Feb 21, 2012 12:43 pm
 Location: Iowa
Re: When to self insure for things like cars? (prefer mathematical model)
I generally carry comprehensive for the first 5 to 8 years of the vehicle. It has nothing to do with my net worth.
Re: When to self insure for things like cars? (prefer mathematical model)
It might go against the conventional wisdom, but I would say comprehensive/collision is a smart move almost regardless of networth. Comprehensive/collision insurance tends to be priced cheaply relative to payouts, with loss ratios near 100%, unlike say umbrella or title insurance. The one exception is if you drive a beater. There's no point in getting collision on a $4k vehicle with a $2k deductible.
 Brianmcg321
 Posts: 986
 Joined: Mon Jul 15, 2019 8:23 am
Re: When to self insure for things like cars? (prefer mathematical model)
All of my cars only have liability. But all three together are less than $10k.
Rules to investing: 
1. Don't lose money. 
2. Don't forget rule number 1.

 Posts: 115
 Joined: Thu Aug 27, 2015 5:14 pm
Re: When to self insure for things like cars? (prefer mathematical model)
For property loss, I would only retain a maximum of 5% to 10% of my expected annual income (labor + investment). This means, if I have a 10k car, I won't insure the property if I have more than 100k to 200k annual income.
The point is that you don't want to disrupt your investment growth and spending too much in case there is a loss event. If you insist, the math model here is a parameterized one: how much loss of annual income you would be willing to forego to not disrupt your daily life and investment growth.
The point is that you don't want to disrupt your investment growth and spending too much in case there is a loss event. If you insist, the math model here is a parameterized one: how much loss of annual income you would be willing to forego to not disrupt your daily life and investment growth.
Re: When to self insure for things like cars? (prefer mathematical model)
What about annual loss of insurance premium?investing engineer wrote: ↑Wed Jul 22, 2020 9:21 pm For property loss, I would only retain a maximum of 5% to 10% of my expected annual income (labor + investment). This means, if I have a 10k car, I won't insure the property if I have more than 100k to 200k annual income.
The point is that you don't want to disrupt your investment growth and spending too much in case there is a loss event. If you insist, the math model here is a parameterized one: how much loss of annual income you would be willing to forego to not disrupt your daily life and investment growth.
Re: When to self insure for things like cars? (prefer mathematical model)
reln wrote: ↑Wed Jul 22, 2020 8:38 pm"Self insurance" is an oxymoron.adherenceEnergy wrote: ↑Wed Jul 22, 2020 7:37 pm Let me give you an example. Let's say you had 10k to your name and your had a 15k car, buying comprehensive and collision insurance should be a no brainer. Now let's say you had 100 million, obviously you wouldn't need insurance on a 15k car, you'd just write a check without blinking. What is the threshold on when you should go without insurance?
My gut instinct is that if your net worth is 10x the value of what you're insuring, you should self insure, but it's not really based on anything. Does anyone know of any mathematical models of when to self insure for items based on your current net worth?
Not really. There are several states that allow you to place a surety bond in lieu of liability insurance. It’s not a good solution for most people. It is an option for some businesses.
Re: When to self insure for things like cars? (prefer mathematical model)
How much the comp and collision costs is pretty germane. If the comp and collision cost $1000 per year and the car is worth $10k I would be way more likely to skip it than if it was $100.
Theoretically, not having it could go a long way towards the purchase of something different. My kid totaled my car in 2019. I had bought it new in 2004. I had full coverage til circa 2011. By 2019 I had probably saved enough by skipping comp and collision to buy a replacement vehicle.
Theoretically, not having it could go a long way towards the purchase of something different. My kid totaled my car in 2019. I had bought it new in 2004. I had full coverage til circa 2011. By 2019 I had probably saved enough by skipping comp and collision to buy a replacement vehicle.
Re: When to self insure for things like cars? (prefer mathematical model)
Which isn't insurance.jbmitt wrote: ↑Wed Jul 22, 2020 10:32 pmreln wrote: ↑Wed Jul 22, 2020 8:38 pm"Self insurance" is an oxymoron.adherenceEnergy wrote: ↑Wed Jul 22, 2020 7:37 pm Let me give you an example. Let's say you had 10k to your name and your had a 15k car, buying comprehensive and collision insurance should be a no brainer. Now let's say you had 100 million, obviously you wouldn't need insurance on a 15k car, you'd just write a check without blinking. What is the threshold on when you should go without insurance?
My gut instinct is that if your net worth is 10x the value of what you're insuring, you should self insure, but it's not really based on anything. Does anyone know of any mathematical models of when to self insure for items based on your current net worth?
Not really. There are several states that allow you to place a surety bond in lieu of liability insurance. It’s not a good solution for most people. It is an option for some businesses.
Re: When to self insure for things like cars? (prefer mathematical model)
I don't know if folks like us have a mathematical model. Insurance companies probably do. That's why they make so much money. Personally, I drop comprehensive/collision insurance once the value of my vehicle goes below <$10K.
Re: When to self insure for things like cars? (prefer mathematical model)
It also depends on who the driver is. A 17yearold teen boy driver is going to be a lot higher risk on average than a 50 year old driver.mnnice wrote: ↑Wed Jul 22, 2020 10:42 pm How much the comp and collision costs is pretty germane. If the comp and collision cost $1000 per year and the car is worth $10k I would be way more likely to skip it than if it was $100.
Theoretically, not having it could go a long way towards the purchase of something different. My kid totaled my car in 2019. I had bought it new in 2004. I had full coverage til circa 2011. By 2019 I had probably saved enough by skipping comp and collision to buy a replacement vehicle.
Re: When to self insure for things like cars? (prefer mathematical model)
That could work if the accident was your fault but if someone else is at fault and their insurance is not paying for the damage, or they were uninsured, then having your insurance would be useful since your insurance company would deal with collecting from the other driver. There can also be situations where it is not completely clear who was at fault. If there are three or more cars involved it can get very complex to figure out who should pay what. Insurance not only covers the cost of damage but also the legal expenses.adherenceEnergy wrote: ↑Wed Jul 22, 2020 7:37 pm Now let's say you had 100 million, obviously you wouldn't need insurance on a 15k car, you'd just write a check without blinking. What is the threshold on when you should go without insurance?
If you are in a fender bender then you would also need to deal with the body shop and you could also end up paying a lot more than an insurance company adjuster would agree to pay just because you do not know what a repair should cost.
In the scenario that you outlined you might want to raise your deductible but the cost of insurance would be trivial to you and the convenience of just being able to hand the keys to the car over to the insurance company to let them figure it out is worth a lot.
I think the point where you would want to drop the collusion and comprehensive would be when the car value is a lot lower, maybe less then $5,000. Your driving record and the cost of the insurance is also a big factor. If you have a bad driving record and your insurance would be very high then that would also favor self insurance.

 Posts: 115
 Joined: Thu Aug 27, 2015 5:14 pm
Re: When to self insure for things like cars? (prefer mathematical model)
I'm only considering lowprobability lossincurring events, so normally you won't expect the premium to be on the same magnitude of the insured property. For some loss events the premium is higher than for other events, but IMO that's just a manifestation of the probability of the loss. As long as the premium is on an affordable level, I won't take too much of it into consideration for the purpose of deciding whether to insure some property.Starfish wrote: ↑Wed Jul 22, 2020 10:11 pmWhat about annual loss of insurance premium?investing engineer wrote: ↑Wed Jul 22, 2020 9:21 pm For property loss, I would only retain a maximum of 5% to 10% of my expected annual income (labor + investment). This means, if I have a 10k car, I won't insure the property if I have more than 100k to 200k annual income.
The point is that you don't want to disrupt your investment growth and spending too much in case there is a loss event. If you insist, the math model here is a parameterized one: how much loss of annual income you would be willing to forego to not disrupt your daily life and investment growth.
Re: When to self insure for things like cars? (prefer mathematical model)
If you cut your comprehensive and collision coverage out, how long will it take for that savings to double your $100M?adherenceEnergy wrote: ↑Wed Jul 22, 2020 7:37 pm Let me give you an example. Let's say you had 10k to your name and your had a 15k car, buying comprehensive and collision insurance should be a no brainer. Now let's say you had 100 million, obviously you wouldn't need insurance on a 15k car, you'd just write a check without blinking. What is the threshold on when you should go without insurance?
My gut instinct is that if your net worth is 10x the value of what you're insuring, you should self insure, but it's not really based on anything. Does anyone know of any mathematical models of when to self insure for items based on your current net worth?
Re: When to self insure for things like cars? (prefer mathematical model)
I never saw much reason for collision coverage regardless of net worth. The only reason to have it I can think of is to appease your lender.

 Posts: 70
 Joined: Tue Sep 24, 2019 6:11 am
 Location: australia
Re: When to self insure for things like cars? (prefer mathematical model)
a standard way to model it mathematically would be to compute the expected value of the utility of buying insurance and compare it to the expected value of utility of not buying insurance. then you would pick the option that maximises your expected utility. of course the results would depend entirely on how you define utility. https://en.wikipedia.org/wiki/Expected_valueadherenceEnergy wrote: ↑Wed Jul 22, 2020 7:37 pm My gut instinct is that if your net worth is 10x the value of what you're insuring, you should self insure, but it's not really based on anything. Does anyone know of any mathematical models of when to self insure for items based on your current net worth?
for utility you'd need to define some function that could quantify how much utility you gained or lost as a function of some other parameters  e.g. perhaps you could declare your utility to be a function of your net worth. you'd need to figure out what shape that utility function had and how to tune the parameters to model your degree of risk aversion etc. there seems to be a cottage industry of interest in this kind of thing in economics, see e.g. https://en.wikipedia.org/wiki/Risk_aversion
one alternative to expected value could be to analyse the situation using prospect theory: https://en.wikipedia.org/wiki/Prospect_theory
Last edited by pseudoiterative on Thu Jul 23, 2020 6:26 am, edited 1 time in total.

 Posts: 1224
 Joined: Tue Feb 23, 2016 4:24 am
Re: When to self insure for things like cars? (prefer mathematical model)
One step you need to think through is insurance pricing model (which I do not know). The assumption this thread is making is it is acturally fair plus a small premium to the insurance company. That might not be true.
The insurance pieces might be bundled for either marketing purposes so getting collision could be cheaper and a good deal or it might be more expensive to split the market and discourage certain types of drivers.
If you want a mathematical model, you need the to understand the companies model.
If you want a simple rule of thumb, then if you can afford it (whatever that means in a risk utility format) do not insure.
The insurance pieces might be bundled for either marketing purposes so getting collision could be cheaper and a good deal or it might be more expensive to split the market and discourage certain types of drivers.
If you want a mathematical model, you need the to understand the companies model.
If you want a simple rule of thumb, then if you can afford it (whatever that means in a risk utility format) do not insure.
G.E. Box "All models are wrong, but some are useful."

 Posts: 7106
 Joined: Thu Oct 15, 2015 3:53 pm
Re: When to self insure for things like cars? (prefer mathematical model)
There is also a mental impact to covering something like collision/comprehension yourself (big expense). For my 6 year old car it costs $196 for a year of coll/comp. The car is probably worth $12k. A $12k hit is very small part of my net worth but still a big expense. $196/year is not. Even if add it up and may come out ahead without the coverage which is unknowable in advance. My last car I dropped coll/comp when it was worth about $4$5k.
 arcticpineapplecorp.
 Posts: 6212
 Joined: Tue Mar 06, 2012 9:22 pm
Re: When to self insure for things like cars? (prefer mathematical model)
this is what Clark Howard says:
When to drop comprehensive and collision coverage on your vehicle and just keep liability insurance depends largely on the age of your vehicle.
The general rule is: If the cost of comprehensive and collision exceeds 10% of your vehicle’s value, that’s the time to dump it and just have liability coverage. You can determine your vehicle’s value at Edmunds.com, KBB.com or NADA.com.
Let’s say you have a 10yearold vehicle that’s worth only $4,000. The minute that you’re paying north of $400 annually — that’s 10% of $4,000 — for collision and comprehensive, it no longer makes financial sense.
One notable exception to this rule: If there’s no way you could financially cover the loss of your vehicle, forget the math and keep paying for collision and comprehensive.
If you have an older vehicle, it often doesn’t make sense to carry full coverage on it. That’s because, if you have an accident, the car has so little value that you’re not going to get a big, fat check to replace it. Depending on the vehicle’s age and condition, you may be lucky to get a few hundred or even a thousand dollars.
Of course, there’s an exception to every rule. Let’s say you remove your collision and comprehensive and just a few weeks later you total your car.
“In that scenario, my advice would prove to be rotten, because you’ll be liable for the expenses on your own,” Clark says. “Statistically it’s unlikely, but it happens.”
https://clark.com/insurance/liabilityvsfullcoverage/
It's "Stay" the course, not Stray the Course. Buy and Hold works. You should really try it sometime. Get a plan: www.bogleheads.org/wiki/Investment_policy_statement

 Posts: 12664
 Joined: Tue Dec 31, 2013 7:05 am
 Location: 26 miles, 385 yards west of Copley Square
Re: When to self insure for things like cars? (prefer mathematical model)
The math model is too complex for me and excel. I'd need rate tables, car value tables and maybe ballistic trajectory calculations.
Do not forget.....as a car's value drops, the collision and comp rates also drop. I have a car that has a value of about $3k that I bought back after the insurance company totaled it. I keep comp for windshield coverage only. It's like $41 a year and having just replaced a windshield in one of my other cars, I know that this $41 number is right about 10% of the cost of the windshield. Do I expect I'll have to replace the windshield (again) in the $3k car (it was replaced years ago once)? You bet. Between our 5 cars and 4 drivers, we lose a windshield almost once a year. I always wait until the month where inspection is due to reduce the chance of losing a windshield right after it was replaced.
Do not forget.....as a car's value drops, the collision and comp rates also drop. I have a car that has a value of about $3k that I bought back after the insurance company totaled it. I keep comp for windshield coverage only. It's like $41 a year and having just replaced a windshield in one of my other cars, I know that this $41 number is right about 10% of the cost of the windshield. Do I expect I'll have to replace the windshield (again) in the $3k car (it was replaced years ago once)? You bet. Between our 5 cars and 4 drivers, we lose a windshield almost once a year. I always wait until the month where inspection is due to reduce the chance of losing a windshield right after it was replaced.
Bogle: Smart Beta is stupid

 Posts: 1704
 Joined: Tue Mar 07, 2017 4:25 pm
Re: When to self insure for things like cars? (prefer mathematical model)
Um, what ? Most Americans can't come up with a grand much less (insert how much your car is worth). If you have a loan on an automobile those #'s go up.
Re: When to self insure for things like cars? (prefer mathematical model)
You also have to ask yourself at what point you would find it worthwhile to file a claim to get the money you have purchased by having insurance. If the liability coverage is by far the most important part of your auto insurance, would having a claim increase the cost of this insurance by enough to cancel out the money you got from the comp or collision? If you have too many claims, even if not at faulty, your company may drop you altogether and you might have to pay very high rates to find insurance.
We can easily afford to replace the kinds of cars we drive, particularly if we do not have to do both at the same time. Comp and collision might be nice to get the insurance company's help and prices for repairs, but we would have to weigh that against the disincentives to filing a claim at all.
We can easily afford to replace the kinds of cars we drive, particularly if we do not have to do both at the same time. Comp and collision might be nice to get the insurance company's help and prices for repairs, but we would have to weigh that against the disincentives to filing a claim at all.
We don't know how to beat the market on a riskadjusted basis, and we don't know anyone that does know either 
Swedroe 
We assume that markets are efficient, that prices are right 
Fama
Re: When to self insure for things like cars? (prefer mathematical model)
This. At a certain point, the insurance premiums become negligible relative to your net worth unless the guy worth $100M drives a Bugatti.shess wrote: ↑Thu Jul 23, 2020 12:24 amIf you cut your comprehensive and collision coverage out, how long will it take for that savings to double your $100M?adherenceEnergy wrote: ↑Wed Jul 22, 2020 7:37 pm Let me give you an example. Let's say you had 10k to your name and your had a 15k car, buying comprehensive and collision insurance should be a no brainer. Now let's say you had 100 million, obviously you wouldn't need insurance on a 15k car, you'd just write a check without blinking. What is the threshold on when you should go without insurance?
My gut instinct is that if your net worth is 10x the value of what you're insuring, you should self insure, but it's not really based on anything. Does anyone know of any mathematical models of when to self insure for items based on your current net worth?
Re: When to self insure for things like cars? (prefer mathematical model)
I'm not sure I would be looking for a mathematical model for this sort of thing. To me, that implies a monte carlo simulation, discounted cash flow analysis, or something. It's just a matter of the consequences to you if you needed to replace your car (or whatever) on short notice. This basically comes down to liquidity. I wouldn't consider a simple net worth, because some assets are illiquid (eg. retirement accounts, illiquid taxable investments). Also, some sources of liquidity aren't going to appear on your balance sheet. Here are some examples of liquidity that may be available:adherenceEnergy wrote: ↑Wed Jul 22, 2020 7:37 pm Let me give you an example. Let's say you had 10k to your name and your had a 15k car, buying comprehensive and collision insurance should be a no brainer. Now let's say you had 100 million, obviously you wouldn't need insurance on a 15k car, you'd just write a check without blinking. What is the threshold on when you should go without insurance?
My gut instinct is that if your net worth is 10x the value of what you're insuring, you should self insure, but it's not really based on anything. Does anyone know of any mathematical models of when to self insure for items based on your current net worth?
 If your credit is excellent and you'd have no trouble getting a lowinterest car loan, maybe you don't need any liquidity because you could just go finance a car. (I know that's not normal BH advice, but we're talking a lowprobability situation.)
 Cash flow (you may have enough float in your checking account and/or monthly surplus to cover an expense in a short amount of time)
 Cash savings (savings accounts, money markets, etc.)
 Taxable investments (these may have tax consequences if they have unrealized gains)
 Borrowing against taxable investments (may be a better way to tap investments with unrealized gains)
 Borrowing against home equity (usually HELOC or cashout refinance)
 Withdrawing contributions from a Roth IRA (again, loss of future taxfree growth is not ideal, but a lowprobability situation)
 HSA withdrawals using saved receipts (although there could be liquidity issues if the HSA is invested in volatile investments like stocks, and there's also a loss of future taxfree growth)
 Hardship payout of accrued vacation from your employer
 Credit card 0% promotions
 Family or friends
As others have stated, liability insurance is a completely separate issue. The potential losses are much higher than collision, and can run well into the 7figure range. If you have enough money to cover a judgment of that size without making a significant dent in your finances, you shouldn't be taking financial advice from me
Re: When to self insure for things like cars? (prefer mathematical model)
Auto insurance bills are usually itemized. Just subtract all the line items from the total premium. Whatever remains is liability.Broken Man 1999 wrote: ↑Wed Jul 22, 2020 8:18 pm Maybe a boglehead who sells car insurance could educate us so far as what percentage the liability portion represents in the premium.
On average, liability is 60% of the auto insurance premium. In 2018, the auto insurance industry earned $243 billion in premiums for private passenger cars in the United States. $146 billion of this was for liability.
Re: When to self insure for things like cars? (prefer mathematical model)
The loss ratio for auto insurance is 65%.oldfort wrote: ↑Wed Jul 22, 2020 8:56 pm Comprehensive/collision insurance tends to be priced cheaply relative to payouts, with loss ratios near 100%, unlike say umbrella or title insurance. The one exception is if you drive a beater. There's no point in getting collision on a $4k vehicle with a $2k deductible.
Collision insurance is a bad deal for a lowvalue car, especially with a $2000 deductible. The fixed fee could be more than half of the collision premium.

 Posts: 5047
 Joined: Wed Apr 08, 2015 11:31 am
 Location: West coast of Florida, inland on high ground!
Re: When to self insure for things like cars? (prefer mathematical model)
I should have worded the post better, I was actually asking about the overall portion for liability, industry data.talzara wrote: ↑Thu Jul 23, 2020 1:08 pmAuto insurance bills are usually itemized. Just subtract all the line items from the total premium. Whatever remains is liability.Broken Man 1999 wrote: ↑Wed Jul 22, 2020 8:18 pm Maybe a boglehead who sells car insurance could educate us so far as what percentage the liability portion represents in the premium.
On average, liability is 60% of the auto insurance premium. In 2018, the auto insurance industry earned $243 billion in premiums for private passenger cars in the United States. $146 billion of this was for liability.
Despite my poor wording, you answered my question!
Thanks!
Broken Man 1999
“If I cannot drink Bourbon and smoke cigars in Heaven then I shall not go. " Mark Twain
Re: When to self insure for things like cars? (prefer mathematical model)
People worth tens of millions are the ones who most need the highest coverage they can get. Lawyers are salivating over the opportunity to sue them. There was a guy in Florida that settled for almost $50 million in a car accident that resulted in death.michaeljc70 wrote: ↑Wed Jul 22, 2020 8:06 pm Unless you have tens of millions, I wouldn't consider self insuring (normal thingsnot counting getting collision on a $3k car).
Re: When to self insure for things like cars? (prefer mathematical model)
There is no crosssubsidy. Insurance companies calculate separate loss ratios for bodily injury liability, property damage liability, collision, comprehensive, PIP, etc. The rates are adjusted separately for each peril to reach the target loss ratio.qwertyjazz wrote: ↑Thu Jul 23, 2020 5:35 am One step you need to think through is insurance pricing model (which I do not know). The assumption this thread is making is it is acturally fair plus a small premium to the insurance company. That might not be true.
The insurance pieces might be bundled for either marketing purposes so getting collision could be cheaper and a good deal or it might be more expensive to split the market and discourage certain types of drivers.
Even homeowners insurance isn't priced as a bundle anymore, except at State Farm. Most of the insurance industry moved to perperil pricing for homeowners insurance 1020 years ago. You're actually paying separate premiums for fire, hurricane, liability, wind, etc. However, it's not itemized on your bill.
Re: When to self insure for things like cars? (prefer mathematical model)
It is certainly interesting to know the loss ratio across the industry on auto insurance.
For an individual, the decision on comp and collision depends on the things we have been discussing ability to pay for the costs of repairs or a new car, size of deductible one would choose, willingness to submit a claim for the amount at stake.
For those who have expensive cars, the insurance company could be on the hook for tens of thousands of dollars, more if you have an S class or exotic car. For those collision and comprehensive may make sense, even if the people who own such cars have the finances to go out and buy another if one were totaled.
For those who drive inexpensive cars and keep them a long time, the payment from the company for the damage to the car, even totaled, many not be much. I suppose one strategy might be very high deductible and keep both coverages for a totaled car or to get the company involved on your side to get compensation from the other driver.
I don't think our two cars combined would cost $10,000 to replace, so not much motive to pay a significant premium. Not having to deal with some careless driver myself, now that would be worth paying for.
For an individual, the decision on comp and collision depends on the things we have been discussing ability to pay for the costs of repairs or a new car, size of deductible one would choose, willingness to submit a claim for the amount at stake.
For those who have expensive cars, the insurance company could be on the hook for tens of thousands of dollars, more if you have an S class or exotic car. For those collision and comprehensive may make sense, even if the people who own such cars have the finances to go out and buy another if one were totaled.
For those who drive inexpensive cars and keep them a long time, the payment from the company for the damage to the car, even totaled, many not be much. I suppose one strategy might be very high deductible and keep both coverages for a totaled car or to get the company involved on your side to get compensation from the other driver.
I don't think our two cars combined would cost $10,000 to replace, so not much motive to pay a significant premium. Not having to deal with some careless driver myself, now that would be worth paying for.
We don't know how to beat the market on a riskadjusted basis, and we don't know anyone that does know either 
Swedroe 
We assume that markets are efficient, that prices are right 
Fama
Re: When to self insure for things like cars? (prefer mathematical model)
Please explain. I would have thought that all insurance companies calculate the risks from the major perils and use those figures to generate a premium. What is the difference between doing this but not itemizing the bill and doing whatever State Farm does? Does SF not account for each of these risks in pricing a policy?talzara wrote: ↑Thu Jul 23, 2020 1:32 pmThere is no crosssubsidy. Insurance companies calculate separate loss ratios for bodily injury liability, property damage liability, collision, comprehensive, PIP, etc. The rates are adjusted separately for each peril to reach the target loss ratio.qwertyjazz wrote: ↑Thu Jul 23, 2020 5:35 am One step you need to think through is insurance pricing model (which I do not know). The assumption this thread is making is it is acturally fair plus a small premium to the insurance company. That might not be true.
The insurance pieces might be bundled for either marketing purposes so getting collision could be cheaper and a good deal or it might be more expensive to split the market and discourage certain types of drivers.
Even homeowners insurance isn't priced as a bundle anymore, except at State Farm. Most of the insurance industry moved to perperil pricing for homeowners insurance 1020 years ago. You're actually paying separate premiums for fire, hurricane, liability, wind, etc. However, it's not itemized on your bill.
We don't know how to beat the market on a riskadjusted basis, and we don't know anyone that does know either 
Swedroe 
We assume that markets are efficient, that prices are right 
Fama
Re: When to self insure for things like cars? (prefer mathematical model)
The fact that most Americans don’t have enough savings doesn’t have anything to do with my view on collision car insurance. I do agree that folks should live within their means and have reasonably sized emergency funds though.deltaneutral83 wrote: ↑Thu Jul 23, 2020 10:01 amUm, what ? Most Americans can't come up with a grand much less (insert how much your car is worth). If you have a loan on an automobile those #'s go up.
Re: When to self insure for things like cars? (prefer mathematical model)
Mostly older vehicles here as well, comprehensive ($100 deductible) costs ~$5/month on each.Jack FFR1846 wrote: ↑Thu Jul 23, 2020 9:10 am The math model is too complex for me and excel. I'd need rate tables, car value tables and maybe ballistic trajectory calculations.
Do not forget.....as a car's value drops, the collision and comp rates also drop. I have a car that has a value of about $3k that I bought back after the insurance company totaled it. I keep comp for windshield coverage only. It's like $41 a year and having just replaced a windshield in one of my other cars, I know that this $41 number is right about 10% of the cost of the windshield. Do I expect I'll have to replace the windshield (again) in the $3k car (it was replaced years ago once)? You bet. Between our 5 cars and 4 drivers, we lose a windshield almost once a year. I always wait until the month where inspection is due to reduce the chance of losing a windshield right after it was replaced.
So I never drop comprehensive coverage even for those vehicles where I no longer carry collision.
Comprehensive coverage paid for a ~$3,000 repair after a deer strike last year.
I usually drop collision when the vehicle's value drops to ~$5,000.
Last edited by ncbill on Fri Jul 24, 2020 4:37 pm, edited 1 time in total.
Re: When to self insure for things like cars? (prefer mathematical model)
State Farm only calculates two premiums for each homeowners policy: an allperils premium and a catastrophe premium. It does not look at losses from each peril. It sets rates based on its total losses from all causes.afan wrote: ↑Thu Jul 23, 2020 1:41 pmPlease explain. I would have thought that all insurance companies calculate the risks from the major perils and use those figures to generate a premium. What is the difference between doing this but not itemizing the bill and doing whatever State Farm does? Does SF not account for each of these risks in pricing a policy?talzara wrote: ↑Thu Jul 23, 2020 1:32 pm Even homeowners insurance isn't priced as a bundle anymore, except at State Farm. Most of the insurance industry moved to perperil pricing for homeowners insurance 1020 years ago. You're actually paying separate premiums for fire, hurricane, liability, wind, etc. However, it's not itemized on your bill.
Because State Farm cannot adjust pricing separately for each peril, its rates are less accurate than those at other insurance companies.
For example, other insurers give you a discount on your theft premium if you install a burglar alarm. State Farm gives you a discount on your allperils premium. If you increase your liability coverage from $300k to $1 million, then you'll get a discount on the additional $700k of liability coverage.
State Farm avoids adverse selection by playing a different game than the other insurance companies. State Farm has much larger multipolicy discounts than other insurers. It also offers a loyalty discount, even though other insurers offer a new customer discount (loyalty penalty). If you have to buy auto insurance and stay 10 years with State Farm, then it's much harder to take advantage of a mispriced homeowners policy.

 Posts: 111
 Joined: Thu Apr 11, 2019 4:28 pm
Re: When to self insure for things like cars? (prefer mathematical model)
Mathematically this would be trying to divide by 0 imo.adherenceEnergy wrote: ↑Wed Jul 22, 2020 7:37 pm Let me give you an example. Let's say you had 10k to your name and your had a 15k car, buying comprehensive and collision insurance should be a no brainer. Now let's say you had 100 million, obviously you wouldn't need insurance on a 15k car, you'd just write a check without blinking. What is the threshold on when you should go without insurance?
My gut instinct is that if your net worth is 10x the value of what you're insuring, you should self insure, but it's not really based on anything. Does anyone know of any mathematical models of when to self insure for items based on your current net worth?
I assume whoever hits me will have no/ minimal insurance, have no assets and immune to lawsuits, and will inflict $,$$$,$$$ in medical costs
Re: When to self insure for things like cars? (prefer mathematical model)
If that assessment could be verified it would be an important input into the 'mathematical model' but any such model would also have an input for risk aversion which would be important in various cases...thus make the model fairly useless.oldfort wrote: ↑Wed Jul 22, 2020 8:56 pm It might go against the conventional wisdom, but I would say comprehensive/collision is a smart move almost regardless of networth. Comprehensive/collision insurance tends to be priced cheaply relative to payouts, with loss ratios near 100%, unlike say umbrella or title insurance.
But, even without verifying that I've kept $2500 deductible comprehensive on my BMW M2 although it's now 2 yr old and replacing it uninsured would be irksome but not a financial challenge. Eyeballing the premium, it does not seem to assume I go wild with the car; it's basically late middle age guy w/ spotless driving record kind of premium, seems like. But I do push that car somewhat hard on winding roads on road trips, not crazily but much more aggressively than I drive say our SUV (which is 15 yrs old so obviously no comprehensive). Wrapping it around a tree is neither planned nor expected, but isn't just 'not gonna happen' as I'd view a single car accident in my sedate driving mode, which is also how I drive the M2 in traffic. Also with comp coverage you're in the (Geico in our case) 'system', report the claim, go to their preferred shop, etc. That was very smooth process when our former Hyundai got damaged by other 100% at fault uninsured driver (us legally stopped at a light) and freak accident with our previous BMW with us not in the car. That's worth something v purely being on your own.
Re: When to self insure for things like cars? (prefer mathematical model)
Since it doesn’t seem like a bunch of folks are considering this, be aware that every time you have a claim it gets noted in your file and has a very high chance of affecting your premiums. If you hold comprehensive thinking you can just use it to replace your windshield a couple times and get your moneys worth, you may end up paying that back many times over for many years with increased premiums.
Re: When to self insure for things like cars? (prefer mathematical model)
On a related note, what does everything think of property damage liability insurance. Do you go above the state minimum and to what amount? The average property damage claim is only $3,000 to $4,000.
Re: When to self insure for things like cars? (prefer mathematical model)
Agree. Worrying about the cost of insurance is only for those who have plenty of time to calculate the small amount of money saved with self insurance if one never has a loss.oldfort wrote: ↑Thu Jul 23, 2020 10:45 amThis. At a certain point, the insurance premiums become negligible relative to your net worth unless the guy worth $100M drives a Bugatti.shess wrote: ↑Thu Jul 23, 2020 12:24 amIf you cut your comprehensive and collision coverage out, how long will it take for that savings to double your $100M?adherenceEnergy wrote: ↑Wed Jul 22, 2020 7:37 pm Let me give you an example. Let's say you had 10k to your name and your had a 15k car, buying comprehensive and collision insurance should be a no brainer. Now let's say you had 100 million, obviously you wouldn't need insurance on a 15k car, you'd just write a check without blinking. What is the threshold on when you should go without insurance?
My gut instinct is that if your net worth is 10x the value of what you're insuring, you should self insure, but it's not really based on anything. Does anyone know of any mathematical models of when to self insure for items based on your current net worth?
Re: When to self insure for things like cars? (prefer mathematical model)
$50 million dollar lawsuit settlements are rare. $50 million dollar lawsuit settlements in private auto accidents are about as rare as winning the Powerball lottery.sk2101 wrote: ↑Thu Jul 23, 2020 1:22 pmPeople worth tens of millions are the ones who most need the highest coverage they can get. Lawyers are salivating over the opportunity to sue them. There was a guy in Florida that settled for almost $50 million in a car accident that resulted in death.michaeljc70 wrote: ↑Wed Jul 22, 2020 8:06 pm Unless you have tens of millions, I wouldn't consider self insuring (normal thingsnot counting getting collision on a $3k car).