4 ms·

Is there a service to show the map of insurance premiums for different disaster types?

by monort 9y ago

sverige 9y ago
Historical loss data is definitely collected by insurance companies for analysis by their underwriters. Sometimes they share that information with each other. For example, some limited loss data is shared among companies for auto claim information associated with auto policy holders. (Not all companies participate. GEICO is the biggest holdout, last I knew.) I have not heard that companies share property damage loss information with each other, but they may do so the same way they share auto claim information. On the other hand, not sharing may give them a competitive advantage. For example, Allstate systematically reduced their coverage of coastal homes in the New Jersey / New York markets about 10 years ago, because their threat models indicated that the area was due for a major hurricane. This obviously reduced their losses when Hurricane Sandy hit. Katrina got all of the big insurers to take a closer look at their exposures from hurricanes, and I would not be surprised to learn that premiums for coastal properties all along the East Coast and Gulf Coast have skyrocketed in the last decade. The risk is very high. They enjoyed over 10 years of low losses, but then this year they got hammered in Texas and Florida. So, would companies share detailed information with each other? Probably not, but they probably have a good idea of what their competitors have paid out for major events, and they definitely know where the risks are geographically. Are they going to share that detailed information with the public? Hell no. They bought it, they paid for it, it's theirs, and they're going to keep it. The easiest way to determine if there's a high risk is to ask for a quote on a property before you buy it. That will tell you a lot. For example, if you're thinking of buying some property in southern California and wonder if the fire risk is high, just ask for a quote. If wildfire risk is high, Allstate (for example) will simply tell you they can't write a policy for that property. Other companies may write a policy, but the premium will be higher than surrounding areas with lower risk. Flood insurance is a special case. No homeowners policy offers coverage for flooding, because it is the textbook example for adverse selection in property and casualty insurance. (That is, only people who live in a flood plain will buy flood insurance, so the risk is not spread across a large enough population. Fire, on the other hand, can happen anywhere, so everyone buys coverage.) For this reason, the government offers flood insurance for those who live in flood plains. Interestingly enough, even then not many people buy the coverage. Edit: Earthquakes are another example of adverse selection. If you live in California and have investigated premiums for earthquake coverage, you already know this. No, it is not covered by your homeowners policy. No, I don't want to pay for it when it happens, because I live in an area with extremely low risk for earthquakes (and flooding). And yet, when it happens, I will surely be paying more taxes to cover someone else's losses.
- monort 9y ago
  Surely this data can be crowdsourced from end-users? Just like wages are not public, but we still have a glassdoor.
  - toomuchtodo 9y ago
    If you could data mine from public record sources, possibly. But there's no way to crowdsource this data due to lack of incentives.

Is there a service to show the map of insurance premiums for different disaster types?

by monort 9y ago

Is there a service to show the map of insurance premiums for different disaster types?

Historical loss data is definitely collected by insurance companies for analysis by their underwriters. Sometimes they share that information with each other. For example, some limited loss data is shared among companies for auto claim information associated with auto policy holders. (Not all companies participate. GEICO is the biggest holdout, last I knew.) I have not heard that companies share property damage loss information with each other, but they may do so the same way they share auto claim information. On the other hand, not sharing may give them a competitive advantage. For example, Allstate systematically reduced their coverage of coastal homes in the New Jersey / New York markets about 10 years ago, because their threat models indicated that the area was due for a major hurricane. This obviously reduced their losses when Hurricane Sandy hit. Katrina got all of the big insurers to take a closer look at their exposures from hurricanes, and I would not be surprised to learn that premiums for coastal properties all along the East Coast and Gulf Coast have skyrocketed in the last decade. The risk is very high. They enjoyed over 10 years of low losses, but then this year they got hammered in Texas and Florida. So, would companies share detailed information with each other? Probably not, but they probably have a good idea of what their competitors have paid out for major events, and they definitely know where the risks are geographically. Are they going to share that detailed information with the public? Hell no. They bought it, they paid for it, it's theirs, and they're going to keep it. The easiest way to determine if there's a high risk is to ask for a quote on a property before you buy it. That will tell you a lot. For example, if you're thinking of buying some property in southern California and wonder if the fire risk is high, just ask for a quote. If wildfire risk is high, Allstate (for example) will simply tell you they can't write a policy for that property. Other companies may write a policy, but the premium will be higher than surrounding areas with lower risk. Flood insurance is a special case. No homeowners policy offers coverage for flooding, because it is the textbook example for adverse selection in property and casualty insurance. (That is, only people who live in a flood plain will buy flood insurance, so the risk is not spread across a large enough population. Fire, on the other hand, can happen anywhere, so everyone buys coverage.) For this reason, the government offers flood insurance for those who live in flood plains. Interestingly enough, even then not many people buy the coverage. Edit: Earthquakes are another example of adverse selection. If you live in California and have investigated premiums for earthquake coverage, you already know this. No, it is not covered by your homeowners policy. No, I don't want to pay for it when it happens, because I live in an area with extremely low risk for earthquakes (and flooding). And yet, when it happens, I will surely be paying more taxes to cover someone else's losses.

monort 9y ago
Surely this data can be crowdsourced from end-users? Just like wages are not public, but we still have a glassdoor.
- toomuchtodo 9y ago
  If you could data mine from public record sources, possibly. But there's no way to crowdsource this data due to lack of incentives.