There are a lot of interesting dynamics in this market.
For example, CAT bonds are generally tied to the specific natural hazard ("this bond triggers if a hurricane of Category 3 or higher land falls in this segment of Florida") or to industry losses, as estimated by an agreed upon source.
This means that a CAT bond is correlated with, but not directly informed by an insurer's actual loss experience. Traditional reinsurance (so an insurer themselves getting insurance) will usually be tied to specific policies, so their experienced loss is what determines payout.
However, depending on the insurer's policies, traditional reinsurance may be unavailable or much too expensive (either due to the large limit needed, the risk level of the policies, or any number of other reasons). Depending on the trigger, a CAT bond can also pay out faster because you don't have to wait to see the claims from 100k home insurance policies.
From the technical side, most large reinsurers license CAT modeling software from one or both of the same two vendors: Moody's RMS or Verisk. The biggest reinsurers will develop their own models, and there
are other modeling vendors that they may license for particular perils (EQEcat for earthquake and KatRisk for flood come to mind), but the big two are pretty widely accepted in reinsurance markets.
That means if your policies are "odd" in some way (uncommon construction type, power facility, etc.), depending on how a reinsurers chooses to model them (or how the model specifically handles them) can have a big impact on your reinsurance pricing. If you know something about your policies that can't be incorporated into a vendor model very well, you may get better pricing on a CAT bond.
These are just some of the considerations! There are so many more things that go into it. But I think it's super interesting to think about.
Source: I work in this side of the industry, specifically in natural catastrophe modeling.
Correct on every point and great insight, hello fellow insurance person.
I will clarify that CAT bonds can have industry loss triggers OR actual indemnity triggers. If an indemnity trigger then the insurer has to prove the actual loss. But you’re right on ILWs (Industry Loss Warranty) in that there is additional model/basis risk considerations.
Insurance companies try to minimize this basis risk. Because while sure it’s great to be in the situation where your CAT bond recovers when you didn’t have large losses, it’s NOT good to be in the position where you had big losses and you don’t recover. Certainty of recover can affect things like how much regulatory credit you get for your reinsurance.
Curious what major trends you’re seeing in your line of work. Guessing global warming induced increase in weather related hazards in some places - anything that would be surprising to people outside the industry?
Having worked on reinsurance software in the 90s, one question that springs to mind, which came to light from the asbestos claims era, was brokers commission. What did happen was brokers would package up risks and sell those off (taking commission) which would see other brokers bundle those up and again package them and others up into a bundle and sell those off. So when a claim came down the line, that was huge, like the asbestos claim in period https://en.wikipedia.org/wiki/Lloyd%27s_of_London which saw such a diluted risk and brokers commission leaching all profit, brought many down financially due to exposure.
So interested how things are today regarding brokers endlessly packaging up risks they sell on, rinse repeat. I'm aware of certain changes that came about to reinsurance brokers in both the Lloyds and London Markets on the back of the asbestos claim era, but not sure of the the CAT model risks/insurance regarding brokers endlessly packaging up to offset risk exposure vs regulations limiting how much they can do that - more so the USA market.
So curious - is there a risk from brokers diluting risks for commission profits in this market or is that saftly covered against and regulated?
> CAT bonds are generally tied to the specific natural hazard ("this bond triggers if a hurricane of Category 3 or higher land falls in this segment of Florida") or to industry losses, as estimated by an agreed upon source.
A big part of the reason for this is that if payout is tied to actual loss, it starts to look a lot like actual insurance, which is specifically not what you want. Because while anyone* can buy a bond, only insurers or reinsurers can write insurance. This is something that needs to be considered whenever an insurer (or anyone, really) tries to transfer risk to a non-insurer.
I've asked two financial advisors about CAT bonds. One had never heard of them and the other said were about as risky as crypto. I guess this is such a niche product that there isn't widespread knowledge about it.
I wonder how much more diversified $ILS could be if it were larger. Would a 10x increase in assets under management give it significantly less volatility because it could do a better job spreading risk around the globe?
The lack of information was my inspiration for building Riskvest. I called my own broker and when I said catastrophic bonds they asked if I meant buying bonds already in default.
On the risk side - your comments here are part of the myth I’m trying to dispel and will have lots more to say in future posts.
Yes for a single CAT bond you are exposed to potential 100% principle losses. But if you buy a bundle of CAT bonds that focus on say California Earthquake, Florida Hurricane, Japanese Typhoon, and a Cyber Event, you can imagine the diversification benefit you get there.
I’ve already created a very very simple model for people to play around with and learn the intuition for CAT bond return patterns. A default means 100% loss and this is unique vs. other bonds. I plan in the future to build a much more robust model.
With the way things are going with climate change, I think that assuming “extreme climate events in different geographies are independent statistical events” is an extremely flawed assumption to make. You acknowledge that it is a simplified model, but that is not some minor oversight. Any model that does not account for this is deeply flawed, and I think no insurance company would choose to model extreme event risk like that. There are common factors (e.g. global average temperature) that can cause many of these events to be triggered in a correlated way.
A “2% risk of default” on an individual bond is something a retail investor might be able to understand, but no one should be buying a “diversified” bundle of these things if they cannot form a reasonable understanding of how correlated they are. Why should understanding the correlation risk be left up to individual investors building their own portfolios?
I also think forming an intuition for these more “all-or-nothing” type events is more difficult than e.g. understanding that if GOOG goes down 10% then AAPL might do too at the same time because they are both tech stocks.
> But if you buy a bundle of CAT bonds that focus on say California Earthquake, Florida Hurricane, Japanese Typhoon, and a Cyber Event, you can imagine the diversification benefit you get there.
Yeah, but imagine how bad a day you're having if all of those disasters happen at once, and then as a cherry on top you lose all your money.
You're not incorrect, but this is the same sort of risk you take when buying an index fund, just that index funds have 100x more entries, so are much more diversified. Eg, we could rewrite this about an index fund like:
"Yeah, but imagine how bad a day you're having if all of those [stocks drop] at once, and then as a cherry on top you [enter a recession]."
I'm not saying this is exactly like buying an index fund. I'm very un-knowledgable about CAT bonds. I'm just saying that your criticism holds for _every_ diversified bundle of risks.
Same principle as why many people prefer not to own shares in the company that employs them -- you're already heavily exposed to that specific risk and don't want to add more. If you live in Florida then a hurricane in Florida already might mean financial loss for you if it damages your house, so buying a CAT bond that covers a different thing is more diversified risk: you might get "house is trashed" or "bond is total loss" but at least you probably will not get both at once.
There are institutional funds but generally it’s a small market with very limited retail presence. Schroders has one. (Artemis is a great source of info in the space - niche trade publication.)
It has been growing slowly for the past 25 years. The limited market size is a reflection of the demand by traditional insurers and reinsurers, for alternative sources of capital. This is as it should be.. when traditional players start transferring risk to the capital markets motivated by the fees involved, or cheaper rates (premium), then you really start worrying about moral hazard i.e. ‘bad risks’ getting transferred to investors.
For the web designers here please let me know if you noticed anything amiss. Ive had particular issues getting captchas working so please comment if you run into that issue.
Thanks for sharing your insight. Wouldn't hurt from a proofread. There are some typos / wrong words, that detract from the sense of authority lent by the article. Eg:
"It's clear that we this structure" --> with
"with out those protections in place" --> without
"Investors would be best to limit their exposer to losses beyond their investment" --> exposure
There might have been others, I had to go back and skim to summarize for you.
So, did the Covid 19 pandemic force multiple insurance companies into insolvency?
Also, what does new product development look like for industries like this? How does one search for new financial products? Is it possible for a non-expert to come up with new products in this space?
Are there any books you can recommend for a novice?
While Covid 19 was certainly a "Catastrophe", the market for pandemic insurance in 2020 was minuscule and to my knowledge did not cause any insurer solvency issues. There were a few cases of insurers being instructed to pay out significant claims by the courts on Business Interruption losses which the insurers argued were not covered due to existing policy exclusion.
Product development is usually highly specialized as there are a lot of nuances and frictions within the insurance industry that outsiders may not fully understand. It helps also to be in the industry and have the network with insurers, reinsurers, brokers, etc. This is not at all to suggest there isn't room for clever people to bring innovation to the market though!
Book recs are hard to specify for CAT bonds but for insurance in general:
Against the Gods: The Remarkable Story of Risk - Peter L. Bernstein
The Black Swan - Nassim Nicholas Taleb
On the Brink: How a Crisis Transformed Lloyd's of London - Andrew Duguid
In the end there's always someone left holding the can. Lloyd's of London has underwriters with unlimited liability. Incredibly, a lot of these Names have historically been private individuals. In the 90s a lot of these lost their shirts (and their homes) when they dicovered that it wasn;t just an easy source of passive income. https://www.theguardian.com/money/2000/nov/04/business.perso...
I'm not aware that these currently exist but the concept of reinsurance on reinsurance is not new (it's called retrocession). I will do some digging and see what I can find - thanks for the suggestion.
There are a lot of interesting dynamics in this market.
For example, CAT bonds are generally tied to the specific natural hazard ("this bond triggers if a hurricane of Category 3 or higher land falls in this segment of Florida") or to industry losses, as estimated by an agreed upon source.
This means that a CAT bond is correlated with, but not directly informed by an insurer's actual loss experience. Traditional reinsurance (so an insurer themselves getting insurance) will usually be tied to specific policies, so their experienced loss is what determines payout.
However, depending on the insurer's policies, traditional reinsurance may be unavailable or much too expensive (either due to the large limit needed, the risk level of the policies, or any number of other reasons). Depending on the trigger, a CAT bond can also pay out faster because you don't have to wait to see the claims from 100k home insurance policies.
From the technical side, most large reinsurers license CAT modeling software from one or both of the same two vendors: Moody's RMS or Verisk. The biggest reinsurers will develop their own models, and there are other modeling vendors that they may license for particular perils (EQEcat for earthquake and KatRisk for flood come to mind), but the big two are pretty widely accepted in reinsurance markets.
That means if your policies are "odd" in some way (uncommon construction type, power facility, etc.), depending on how a reinsurers chooses to model them (or how the model specifically handles them) can have a big impact on your reinsurance pricing. If you know something about your policies that can't be incorporated into a vendor model very well, you may get better pricing on a CAT bond.
These are just some of the considerations! There are so many more things that go into it. But I think it's super interesting to think about.
Source: I work in this side of the industry, specifically in natural catastrophe modeling.
Correct on every point and great insight, hello fellow insurance person.
I will clarify that CAT bonds can have industry loss triggers OR actual indemnity triggers. If an indemnity trigger then the insurer has to prove the actual loss. But you’re right on ILWs (Industry Loss Warranty) in that there is additional model/basis risk considerations.
Insurance companies try to minimize this basis risk. Because while sure it’s great to be in the situation where your CAT bond recovers when you didn’t have large losses, it’s NOT good to be in the position where you had big losses and you don’t recover. Certainty of recover can affect things like how much regulatory credit you get for your reinsurance.
Curious what major trends you’re seeing in your line of work. Guessing global warming induced increase in weather related hazards in some places - anything that would be surprising to people outside the industry?
Having worked on reinsurance software in the 90s, one question that springs to mind, which came to light from the asbestos claims era, was brokers commission. What did happen was brokers would package up risks and sell those off (taking commission) which would see other brokers bundle those up and again package them and others up into a bundle and sell those off. So when a claim came down the line, that was huge, like the asbestos claim in period https://en.wikipedia.org/wiki/Lloyd%27s_of_London which saw such a diluted risk and brokers commission leaching all profit, brought many down financially due to exposure.
So interested how things are today regarding brokers endlessly packaging up risks they sell on, rinse repeat. I'm aware of certain changes that came about to reinsurance brokers in both the Lloyds and London Markets on the back of the asbestos claim era, but not sure of the the CAT model risks/insurance regarding brokers endlessly packaging up to offset risk exposure vs regulations limiting how much they can do that - more so the USA market.
So curious - is there a risk from brokers diluting risks for commission profits in this market or is that saftly covered against and regulated?
> CAT bonds are generally tied to the specific natural hazard ("this bond triggers if a hurricane of Category 3 or higher land falls in this segment of Florida") or to industry losses, as estimated by an agreed upon source.
A big part of the reason for this is that if payout is tied to actual loss, it starts to look a lot like actual insurance, which is specifically not what you want. Because while anyone* can buy a bond, only insurers or reinsurers can write insurance. This is something that needs to be considered whenever an insurer (or anyone, really) tries to transfer risk to a non-insurer.
I've asked two financial advisors about CAT bonds. One had never heard of them and the other said were about as risky as crypto. I guess this is such a niche product that there isn't widespread knowledge about it.
I wonder how much more diversified $ILS could be if it were larger. Would a 10x increase in assets under management give it significantly less volatility because it could do a better job spreading risk around the globe?
The lack of information was my inspiration for building Riskvest. I called my own broker and when I said catastrophic bonds they asked if I meant buying bonds already in default.
On the risk side - your comments here are part of the myth I’m trying to dispel and will have lots more to say in future posts.
Yes for a single CAT bond you are exposed to potential 100% principle losses. But if you buy a bundle of CAT bonds that focus on say California Earthquake, Florida Hurricane, Japanese Typhoon, and a Cyber Event, you can imagine the diversification benefit you get there.
I’ve already created a very very simple model for people to play around with and learn the intuition for CAT bond return patterns. A default means 100% loss and this is unique vs. other bonds. I plan in the future to build a much more robust model.
https://www.riskvest.io/data-lab/cat-bond-portfolio-simulato...
With the way things are going with climate change, I think that assuming “extreme climate events in different geographies are independent statistical events” is an extremely flawed assumption to make. You acknowledge that it is a simplified model, but that is not some minor oversight. Any model that does not account for this is deeply flawed, and I think no insurance company would choose to model extreme event risk like that. There are common factors (e.g. global average temperature) that can cause many of these events to be triggered in a correlated way.
A “2% risk of default” on an individual bond is something a retail investor might be able to understand, but no one should be buying a “diversified” bundle of these things if they cannot form a reasonable understanding of how correlated they are. Why should understanding the correlation risk be left up to individual investors building their own portfolios?
I also think forming an intuition for these more “all-or-nothing” type events is more difficult than e.g. understanding that if GOOG goes down 10% then AAPL might do too at the same time because they are both tech stocks.
> But if you buy a bundle of CAT bonds that focus on say California Earthquake, Florida Hurricane, Japanese Typhoon, and a Cyber Event, you can imagine the diversification benefit you get there.
Yeah, but imagine how bad a day you're having if all of those disasters happen at once, and then as a cherry on top you lose all your money.
You're not incorrect, but this is the same sort of risk you take when buying an index fund, just that index funds have 100x more entries, so are much more diversified. Eg, we could rewrite this about an index fund like:
"Yeah, but imagine how bad a day you're having if all of those [stocks drop] at once, and then as a cherry on top you [enter a recession]."
I'm not saying this is exactly like buying an index fund. I'm very un-knowledgable about CAT bonds. I'm just saying that your criticism holds for _every_ diversified bundle of risks.
Yeah, it seems like you’d want to buy bonds that covers areas that you’re not personally in…
Why? It's not like you can influence the trigger of any such catastrophe
Same principle as why many people prefer not to own shares in the company that employs them -- you're already heavily exposed to that specific risk and don't want to add more. If you live in Florida then a hurricane in Florida already might mean financial loss for you if it damages your house, so buying a CAT bond that covers a different thing is more diversified risk: you might get "house is trashed" or "bond is total loss" but at least you probably will not get both at once.
It’s almost like we should have bundled all these bonds as a product instead of selling single bonds.
There are institutional funds but generally it’s a small market with very limited retail presence. Schroders has one. (Artemis is a great source of info in the space - niche trade publication.)
https://www.artemis.bm/ils-fund-managers/schroder-investment...
It has been growing slowly for the past 25 years. The limited market size is a reflection of the demand by traditional insurers and reinsurers, for alternative sources of capital. This is as it should be.. when traditional players start transferring risk to the capital markets motivated by the fees involved, or cheaper rates (premium), then you really start worrying about moral hazard i.e. ‘bad risks’ getting transferred to investors.
CAT bonds are typically restricted to institutional investors. I would be very surprised if you could even buy one without being a QIB.
https://riskvest.io/market-watch/ils-nyse-offering
New opportunities for retail investors have been popping up lately! That’s a big part of what Riskvest is meant to document.
Blown away by the traffic from this post!
For the web designers here please let me know if you noticed anything amiss. Ive had particular issues getting captchas working so please comment if you run into that issue.
Thanks for sharing your insight. Wouldn't hurt from a proofread. There are some typos / wrong words, that detract from the sense of authority lent by the article. Eg:
"It's clear that we this structure" --> with
"with out those protections in place" --> without
"Investors would be best to limit their exposer to losses beyond their investment" --> exposure
There might have been others, I had to go back and skim to summarize for you.
Thank you! No matter how many times I read things over I managed to leave a few behind. I will make those edits shortly.
> For the web designers here please let me know if you noticed anything amiss.
The images some of the visuals do not show in reader view in Safari and Firefox. Other than that, the content is well laid out and very readable.
Very minor nit: "It's clear that we this structure,..."
Cool article, it's a clear explanation of something I never knew about.
So, did the Covid 19 pandemic force multiple insurance companies into insolvency?
Also, what does new product development look like for industries like this? How does one search for new financial products? Is it possible for a non-expert to come up with new products in this space?
Are there any books you can recommend for a novice?
While Covid 19 was certainly a "Catastrophe", the market for pandemic insurance in 2020 was minuscule and to my knowledge did not cause any insurer solvency issues. There were a few cases of insurers being instructed to pay out significant claims by the courts on Business Interruption losses which the insurers argued were not covered due to existing policy exclusion.
Product development is usually highly specialized as there are a lot of nuances and frictions within the insurance industry that outsiders may not fully understand. It helps also to be in the industry and have the network with insurers, reinsurers, brokers, etc. This is not at all to suggest there isn't room for clever people to bring innovation to the market though!
Book recs are hard to specify for CAT bonds but for insurance in general:
Against the Gods: The Remarkable Story of Risk - Peter L. Bernstein
The Black Swan - Nassim Nicholas Taleb
On the Brink: How a Crisis Transformed Lloyd's of London - Andrew Duguid
The last one is a personal favourite of mine
In the end there's always someone left holding the can. Lloyd's of London has underwriters with unlimited liability. Incredibly, a lot of these Names have historically been private individuals. In the 90s a lot of these lost their shirts (and their homes) when they dicovered that it wasn;t just an easy source of passive income. https://www.theguardian.com/money/2000/nov/04/business.perso...
The impact was nuanced, and depended on the specific policies in place.
Some businesses had business interruption insurance which paid out. Many policies exclude highly correlated events such as pandemics.
And then think about specific events which were cancelled, which may have bought policies protecting them if cancelled.
And of course life insurance and health care would have been affected.
SwissRe often produces public reports in the space if of interest:
https://www.swissre.com/risk-knowledge/building-societal-res...
I don't think so. I'm pretty sure it was considered an 'act of G-d', not an act of China :)
I think you could also have specific pandemic insurance, and that paid out, but those were rare before Covid.
Why would it? I don't think that much pandemic insurance is written and obviously you model all the contracts as being very highly correlated.
What a wonderful read! This is why I come to HN: technical, yet approachable, discussions on topics I didn't even know existed. Thank you for sharing!
Great site, very well written and great explanations of insurance industry dynamics.
This feels like when Selena Gomez explained CDOs in The Big Short.
This topic feels closely related to GFC.
(Margot Robbie)
Different scene. Margo Robbie explained sub-prime mortgages. Selena Gomez explained synthetic CDOs.
TIL, thank you!
Would you consider a followup post about reinsurance assets targeted to derisk potential systemic or liquidity risks in the entire CAT bond market?
I'm not aware that these currently exist but the concept of reinsurance on reinsurance is not new (it's called retrocession). I will do some digging and see what I can find - thanks for the suggestion.
Big fan of your content design :)
Thank you! As an Actuary I'm not exactly known for my design skills so it was a lot of effort to get things looking the way I wanted.
[dead]