Off topic but I really hate modern web design. I found the content of this article interesting but I could hardly read it scrolling through in-article ads, banners, etc. One of the reasons I like HN is the prevalence of personal blogs that just have text for me to sit and read.
Reading some of the compendiums with Watersons commentary are a really interesting look into how things changed over the years and an insight into comic publishing
The notion that big business is conspiring to keep children from playing in the rain because of… reasons… might be my new favorite conspiracy. I’m bought in.
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.
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 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.
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.
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.
The lack of quality is what gets to me. I've used AI tools in many aspects of my life to great benefit. Yet, nowadays scrolling through Reddit, X, or even video based platforms are a deluge of drivel. It was bad enough that I was spending too much time on my phone instead of interacting with other people but now even the content I'm interacting with isn't human!
As someone who learned most of my initial coding abilities through R and RStudio in a data science context, and since moved on to more “standard” languages and IDEs, I’ve yet to find anything that comes close to the flexibility and integration of RStudio for hacking together data analytics.
VS Code/Python has made some major improvements in the past couple years but it’s still very clunky compared to the ease of running R code line by line without having to start up a debug instance. And now with copilot the most frustrating parts of R (such as remembering all the Tidyverse syntax) have been abstracted away.
I believe it was Schopenhauer who said in one of his essays that (I’m paraphrasing) just because content is new, doesn’t make it good, because with a constant stream of new content, what’s good doesn’t stay new for long.