State of Decentralized Insurance - Renat Khasanshyn

oh hi just making sure please give it up for renat who's going to tell us about the state of decentralized insurance [Applause] how are you guys doing so great to to be here again uh it's my third in denver it was amazing to be here when it was uh very few people a much smaller group but next few days actually a couple of weeks will be amazing for all of us to reconnect it's really great to to see you in person so i will share with you the uh my take on the state of decentral's insurance on the opportunities on the challenges and what the various teams in the space are working on and how we collectively work on making both the d5 products and um non-defined traditional insurance available and facilitate mass adoption for the d5 for crypto and for the non-crypto use cases i will show you the the state of them how they help various people around the world achieve financial independence with lower risk so uh i will first share with you the state of the non-defy use cases and they're only a couple they're quite hard to implement i had a great pleasure to be part of a team called etherisc which started with the fly delay protection on uh e3 remain net for developers who flew to shanghai to for devcon and the the defcons that happen in cancun in prague we continue to uh work on uh on this product as uh example building block is a composable uh primitive that anyone can use so uh that led to the this current uh product it's up uh and working you can guys use it for flights um and it's being used as a template for projects like um insurance for crops this project in kenya went live and the first kenyan farmers received payouts there were over 17 southern farmers getting insurance we faced a number of challenges to bring it to market one challenge is that the cost of the amounts that are being protected are so small that it's extremely hard to distribute it when you have a total payout of ten dollars right how do you bring it to the hands of people right there is simply not enough unit economics in there to justify the the syllabus so we found some ways to to solve this problem by bundling protection uh with the bag of seeds so when the farmers buy a bag of seeds they could activate the protection that comes with this bag of seed they they don't use metamask they just most of them have phones that are not even smartphones so they just use them pesa through sms message they activate the policy and uh this uh uh this year we're in the second season of these policies being distributed uh i think that in 2022 we'll have maybe 100 southern canyons there is also a live map uh we can see uh where the policy is being bought but this is only for the off-chain uh offline kind of use cases where traditional insurance is it's really hard problem but what majority of the of my rest of my update i will focus on the state of uh crypto native uh insurance for the risks that are native to those of you who are either holding some assets like stable coins or doing yield farming and so on uh all together as you can see the insurance it's very here in the red it's a small tiny portion of total valley locked uh i will explain the reasons why is that and what it will be when uh the some of the biggest problems we're facing today will be solved these are some of the examples of the existing projects in the space that are uh taking on the crypto native risks such as smart contract hacks uh and uh very popular um product uh dpeg of stable coins um this these are the two major categories of risk the third probably major category is the slashing risk so if we compare uh where we are in in in crypto and if we say that we're building um uh web 3 economies that have their own economic participants gdp right and if we look into the world that is non-crypto approximately 10 percent of all worldwide gdp are premiums of insurance in some countries like taiwan hong kong in the u.s the ratio of premiums to gdp is greater in developing countries the ratio are smaller some developing countries have two three four percent of gdp but uh on average it's about eight ten depending who and who calculates using what kind of data but so if we're using it as a benchmark and we say that uh we are building web 3 economies that have economic activity participation assets the it's very hard to measure gdp of web 3 economies right so what can we measure we can look into the total valley log right as a proxy and if you look into how total value looked is growing this by the way chart from a few months ago it's now outdated so we now have about maybe 200 billion dollars in total volley locked but if we compare this on this chart you can see that the insured tvl is not catching up the total value locked in in d5 is growing so rapidly while the the insurance capacity and the products available they are not catching up so let's examine what what might be the reasons why these uh this is the case and what are the bottlenecks that prevent availability of protection which is uh often prerequisite for growing the economy for growing the total value log because there is so much assets uh in the world uh that are uh coming to um two web 3 economies both existing new and it's critical to understand what what are the bottlenecks that prevent mass adoption right and how can risk transfer the magic of risk transfer help facilitate mass adoption because obviously if someone uh can earn not even 20 but even five percent insured yield in d5 right so much better than insured uh 0.1 percent yield in the bank so a lot of assets would would come and have their d5 counterparts right people would obtain stable coins and have their savings uh in in them and and start saving in in crypto and defy however they don't because majority of them don't because they always ask where's the catch right this this stable coin uh earns 20 yield it must there must be something wrong with it right so um if we could get insured yield right an equivalent of fdic insurance that would facilitate mass adoption at a much greater scale so i will show you some of my uh favorite projects uh well i i'm i'm biased i'm working on a on a couple projects at risk and atomica well i will uh maybe share with you at the end of the my deck uh what's there but i want to highlight some of my favorite projects out there perhaps the most advanced in the form of both approach towards uh supporting risks on multiple chains supporting various kinds of risks when it comes to diversification would be insurance it's one of the latest entrants in the space and as you can see it's a cross-chain risk transfer protocol it has a great growth curve for the both amount of covers and amount of capacity and premium so typical risks that are being insured are risks of smart contract hacks stable coin d-pack and slashing right so finance is another one uh it has about as you can see here there is a spartan bucket that's kind of a reinsurance pool right which uh issues uh protection uh to various kinds of policies including smart contracts as you can see here slashing and downtime right and dollar pack stability use tpeg right and as you can see the on the left with amount of capital that is backing all of these risks and policies right and the yield over here is a combination of the natural yield coming from the individual policies and also the uh they're one of the first examples in the defined insurance where they are deploying the capital uh to earn yield while it's idle right so this is the form of what is for traditional insurance companies this is the number one source of earnings right it's called investment income there is underwriting income the premiums and then there is investment income some insurance companies out there that you have auto policies or home policies they're really not making money on underwriting income they're just making money on investing assets that they have so unslashed is a good example of how uh traditional practices of conservative investment of assets for the purpose of generating investment income is is such an example so it's a sign of space maturing and uh let's ask the question why why does this uh space uh remain a giant unsold problem um which most people believe in the space who are working in the in in the risk transfer risk branching industry we believe that we still have a lot of problems to solve before uh protection will become available uh before it will become available in an easy-to-use manner i'll show you some of the bottlenecks and uh some of the symptoms we can say uh are expensive policies right obviously the question is okay how much will it cost and if it costs five ten percent okay it's not that uh that attractive especially if you're just protecting idle funds that are not earning yield it's a different story if you're earning twenty percent yield right and you you can certainly afford five percent um uh insurance premium right but uh we're looking into a great divergence of premium costs of course it's cost of capital um but overall it's still expensive uh the second biggest problem the symptom is that it's not available for 99 of d5 projects if you're just launching something here at the hackathon right you build something you cannot just go to to some website and get your projects insured right typically you will be asked questions okay do have audit and so on right and then you will get a listing if you're lucky and it may take months before protection will be available for your new project or new product uh whether it's financial non-financial so um the third symptom is that uh uh believe it or not but there is no problem with demand for insurance right even at the prices uh that high the demand is enormous it's so enormous that there is uh there is a waiting list of uh of uh buyers who are waiting for policies to become available and they're buying out all the capacity out there the reason why is because for use cases like yield farming right uh if you're earning 10 20 yield it's not a problem to buy a policy and pay five percent insurance right so uh there is so much capital out there that is uh waiting for the policies to become available yeah yes it's a smart contract hacks the number one uh and it's the bag uh protection of a stable coin let's say you buy usdt and you invest it right the more conservative investors out there that have uh billions of dollars in uh capital that they can deploy uh they're asking a question what if usdt is not backed by dollars right what if it will be priced at 70 cents on a dollar right uh either a way for me to get protection so i can uh sleep well and enjoy my five percent yield right so uh the same thing like so smart contract hacks that includes the recent big acts of bridges right like wormhole attack last week there was a meteor io attack also actually this week right uh that happened on like tuesday uh these are all examples of policies that are in high demand but look for meter you could not even buy it's it's not available for meter right for warm hope i'm not sure if there was any anyone wasn't sure but that's the case did it answer your question yes yeah yeah there is a demand the question was is there a demand for price stability of other non-dollar denominated assets it could be for example would be wbtc at some point someone will say well it's more a bit centralized equivalent of btc right so you can have a d-pack for wbtc you can have d-pack for uh solana version of ether right when it comes to this uh wormhole bridge example of attack the question is how can you incentivize amount of i guess capacity if there is so much demand from institutional investors and uh the answer is that uh it's not just in incentives right all the projects mentioned like uh unslashed uh and nexus and some insurance right and everyone who is launching and if they have a token right of course they will try to incentivize unslashed finance for example a portion of that yield that you earn as a capital provider is the usf token that you earn on top of the natural premium underwriting income and investment income right so they're incentivizing it but i will show you the underlying reasons that are fundamental problems that prevent these uh chicken and egg problem from being sold right number one problem is that most of this now we have new new protocols coming in that are that solve this problem but in case of nexus for example mutual right and in case of let's say on slash finance right you you must provide capital in the form of eth for example you wrap it under it's a on the curve but still it's uh capital that is backing it right it's uh ease die right um it's a big problem right you cannot deploy wbtc and be a capital provider right so uh limited choices of the what you can back under right with is the problem second big problem is that humans today decide prices instead of market forces of supply and demand so uh a third problem uh underlying is the um that human decides on listings right uh uh like meter would probably get uh their bridge covered if they could get a quick quick listing like you could go to uni swap create a pool you cannot create insurance for your own project you build here and uh it will take months before you can right because again there is a listing process and then the next underlying problems are that policies must be a certain amount of duration right obviously it requires capital to be locked for this amount and of course that means that that capital is expensive because you're asking people to low capital for like six months for example lack of point of sale integrations meaning that if you investing into some farm making a deposit right you cannot really check the box and have your deposit being insured right you have to go to a separate website right and so on and so forth then capital pool uh conversion to project token and reserve models all together that makes prices high and yield for capital providers low and investors who specialize in defy risks they agree that defy basically smart contract insurance is one of the kinds it's a unsolved problem yet segment where various solutions to these various problems must be implemented before we can have low prices great selection and hackathon participants demonstrating an insured product that they can offer to the masses right example of that as a byproduct of these symptoms is that as you can see the the overall cover mounts right are not growing uh as rapidly as they could should this problem be solved the overall capacity is not that great um because this again this bottlenecks so and sometimes even like in case of nexus mutual that they're trying to get away from but they're still uh requiring uh uh buyers of protection to be kyc'd and uh they must be living in certain countries right if you one of this country uh residents or citizens you're not eligible so all of these uh problems being sold in various ways seems like i'm i'm uh running out of time to to fill my project that don't have it but i'll just show you a a highlight a screenshot uh of how it looks like for depec protection for example if you buy a policy it could be basically push us i'll just show you some screens but uh what's amazing is for uh many people uh what the uh what what's coming is that opportunity to earn in the space without capital so you don't have to have capital to be participants so what we have done in the atomic in particular is uh four ways to earn uh in space without having any capital and these are product delegates market delegates front and operators and syndicate managers these are the rules that are uh we modeled it after the our future state of the decentralized insurance what it takes in order to bring it to the hands of users in the form of checkbox so that all majority of deposits out there will be insured just as they are insured with the fdic so thank you for your time and if you have any quick questions i can take a couple go ahead yeah so the right now uh there is not much of a aiml uh used in um in define insurance in particular uh for crop insurance for fly delay uh yes it's being used uh one of the speakers here uh later at the event jake brookman uh he was like five years ago when we started working on decentralized insurance he wrote python script that used the past history of flights being delayed to price the cost of a policy so it will see more of it but one of the solution to a problem of pricing is is basically make it based on supply and demand and our approach is to price it on utilization curve like compound does uh prices their capacity so for now we we should use much simpler mechanisms than aiml they will improve the pricing but 80 of impact will be done using the automated market making and pull utilization approaches to pricing go ahead yeah it's using