Institutional Investment and Trading Strategies in DeFi - Alex Elkrief

all right so hello everyone my name is alex um i run the book of yield generating strategies for ledger prime it's not as ex as exciting as daos and pfp projects i'm afraid but we'll have to make do for now um so first a little bit of background on ledger prime number one we're not associated with ledger wallets in any way we get a lot of confusion about that a lot um we are a multi-strategy asset management firm we the dna of the firm started in options trading crypto options and evolved into d5 as the space emerged where big liquidity providers on all the options derivatives exchanges thereabit bit.com cme etc and then we've invested in over 50 projects uh over the years in general our value add is that we try to build things for the projects that we invest in whether it's making market market making bots on injective or making markets behind ribbon finance or just providing liquidity to bootstrap the projects that we're uh we're getting invested in we're also recently started a a group called leisure prime labs where we are looking to incubate projects of our own and taking the learnings that we've had as investors over the years and so we're spending a lot of time thinking about both the yield optimization problem from the investor looking to generate returns but also from the side of the d5 builder trying to design solutions that are sustainable for the long term especially on the tokenomics side so i think all in all it's a question of aligning incentives between liquidity providers and um and builders so i'll cover briefly uh the current state of defy certain tokenomic models that have uh that have been prominent over the last year and then where we can we can go from here along several axis of innovation and then fourth point the boring one but probably necessary is uh some pieces of institutional tooling that will make institutions a lot more comfortable to participate in the space so where's d5 um we've seen an 18x tvl increase over the past year from about 18 billion to 240 billion while 2020 was mostly d5 summer was mostly about the simple amm curve unit swap sushi swap and then money market funds money market protocols like compound and ave we saw a lot of innovation in the types of primitives that are created in d5 over the past year i'm listing a few there but there are more options options amms um d5 2.0 and protocol owned liquidity and then yield trenching leverage farming and then bridges that have become pretty critical both in a good and bad way to the space the third point is that d5 is now truly multi-chain whether we like it like it or not there are over 300 protocols across 35 chains that have at least 30 million in tbl we've seen over 2.3 billion in ecosystem development funds being announced and the number i'm quoting here is at the time of the announcement back when phantom was at 50 cents and avalanche was at 30 dollars so i think we can safely multiply that number by two and um so the consequences of that for for projects is that every project is in a constant search for attention from the d5 circus um and attracting capital right and left so that's the side from the d5 builders and from the side of the yield optimizer the the world has also evolved pretty significantly over the past year we've seen the the game of optimization has gone from optimizing between ave and compound to multi-layered strategies on multiple chains you start with your providing liquidity to curve you put that into convex then you put those convex tokens into fracs and then it goes on and on yields are also very volatile and opaque what you see on the on whatever app you're on is rarely what you get there are small pieces of slippage right and left and entry and exit fees and a 25 basis point entry or exit fees is actually non-trivial if you're trying to generate 20 on your position and it gets even worse if you're 20 that you see on the screen today is 23 tomorrow and then 17 the next day and then and keeps going up and down like this and so from the side of the investor more projects more chains you know every chain needs a dex an amm a a compound fork an ave fork they're all throwing money at the problem to incentivize liquidity and so from a liquidity provider you're stuck with the need to rebalance your positions all the time and the cycles of boom and bust um have gotten shorter and shorter so where does that leave us in terms of objectives um i think on the side of d5 projects is how to design tokenomics that incentivize long-term liquidity partners and also moving away from generating yield on the basis of inflationary tokens towards products that are actually generating sustainable yield based on market risk and on the other side liquidity providers would love to just have positions that they can stay in for multiple weeks and forget about without having to you know monitor the next 36 projects that launched last week so i'll cover two types of tokenomic designs that we can learn from from the past year or year and a half the first one is the master the sushi swap master surf contract the current the general structure is that you provide liquidity to a protocol you earn governance tokens there's no lock up on the governance tokens that you earn and then you put that into what is called the pull 2 where you stake those governance tokens for more governance tokens in general it starts with a lot of liquidity mining available which means that people get excited they buy the token you get reflexivity on the upside because the the yield is paid in native tokens so uh higher token price better yield people get more excited et cetera et cetera on the way up the problem is then the music stops at some point and you get the same reflexivity on the way down and so that chart is from a real project it's a pretty typical chart that you see as token launches and then the hype dies down and again the d5 circus moves to the next shiny new thing the second iteration that's become more and more popular is is uh the curve model of vote escrow tokens um the difference here is that you provide liquidity and uh the more governance tokens you stake and the longer you stake them for the more of a boost you get to your yield between one the the luck is between one and four years the problem so it does solve the problem of longer term alignment you you can lock your tokens for four years you're aligned with curve the problem with that is that unless you have permanent capital you have a vc who has a very long life cycle four years is very stringent especially in this environment and defy being permissionless it immediately incentivized convex finance to create synthetic derivatives that effectively give you the boost without having to stake so it's a it's better than the first model but it creates synthetic derivatives and and convex ends up owning over 42 percent of the curve supply today which is not great for the purpose of decentralization so where where can we innovate um there's really three main axes the first one is tokenomics and designing liquidity mining incentives in a better way the second one is moving away from inflationary yield in the form of real financial products that are being replicated on chain paid options exotic derivatives we don't have a lot of interest rate products yet but i expect that will come soon and then capital efficiency the the quantity of assets that are locked in pools that have less than five percent utilization is extremely high and there's this elusive dream of achieving some amount of cross-margining which interestingly enough d5 being permissionless and composable positions it probably better than crypto c5 to achieve that so i'll cover three case studies um one for each of those points the first one on the tokenomics side that i find interesting is platypus similar to curve you provide liquidity to a stable pair imm you earn pdp rewards you can stake those pdp rewards and the more you stake the bigger boost you get on your on your yield the key difference is that you're not locked up there's a time component to it so the longer you stake retroactively the more boost you accumulate but the moment you unstake a single pdp token you lose all of your boost so it does solve the problem of longer term alignment as a liquidity provider you're incentivized to deposit your tokens for as long as possible because you keep accumulating boost but you're not locked up on the primitive case study there are many and i just picked one of them lyra on optimism is a great case lyrisin is a is an options amm where you essentially have a pool of capital against which calls and calls and puts are traded the pool of capital it's not done yet but will eventually automatically delta hedge and so as a liquidity provider or as an investor to the pool you're you're generating yield not there is liquidity mining but that won't last forever but mostly you're generating yield by taking a short volatility position which is more sustainable because it is an actual market risk driven yield third case study on the capital efficiency point is gearbox i think they're doing something really interesting with the notion of composable leverage they've essentially created a system where you come up with your usdc or ether whatever token and you create this concept of a credit account which enables you to lever up but that is just an independent leverage module that then gets integrated with other protocols so you come with your ten thousand dollars you lever that up four times and then you can go and deposit that into curve or yearn that um that is then a lot more four times as capital efficient and so if you if you combine the last three case studies that i've just mentioned you can see where we can we can have the space evolve where if you take lyra that is generating or any real final financial product that is generating real yield with pdp style tokenomics where you're incentivized for the long term you have long term inflation so you can trust that this boost is going to last for a long time and then you can lever up or you can be a lot more capital efficient that starts to be a lot more interesting and capital efficiency will be will be necessary to compete with any traditional financial system which is what we're trying to do here this is the point where a transition to the more boring pieces but that are pretty necessary we're seeing a lot of project launch with only metamask support this excludes a lot of custodians that rely on wallet connect integrations we're still seeing way too much of the space that relies on ledger while it's deploying millions and millions of dollars in defy and that's very scary so if you're a project building d5 out there please integrate while connect and then these are two other points that i don't really have the answer to but the the way contracts are designed right now trigger a lot of taxable events when you claim your tokens that's a taxable event you stake your tokens so you take you stake um sushi for ex-sushi that's effectively a trade and then when you unstake you get ex sushi back for sushi and then you probably sell your sushi that's four taxable events and so it would be great to see some innovation in the space where you could for example directly stream your staking rewards into this taking contract and do everything in as few transactions as possible to limit the tax implications it also applies to the third point any institution that is involved in d5 right now is having a nightmare explaining what they're doing to their fund admins i record every every single transaction that i do and i don't think there's a single fun admin in the world that will go and read my ether scan transactions so again batching and building wrappers around certain strategies in one block would be extremely helpful and then there's a lot of capital that is not yet comfortable with defy and these are some of the reasons why that's not the case you go to a website and you see a yield that's very attractive you have very little transparency on to what the yield has been until now and what is actually driving that yield how governance changes might impact that yield and then the apr the aprs that you're seeing are not standardized at all it's unclear whether it's the last it's averaged over the last 24 hours the last seven days the last 30 days or if it's forward looking and then you go to convex and you have a virtual apr and so having having some amount of standardization there would be extremely helpful so how do we attract long-term capital um stable transparent and predictable liquidity mining campaigns reducing the cycle of boom and bust that is exhausting for everyone building products that generate sustainable yield and by sustainable i mean non-inflation non-token inflation driven and then integrating with more and more institutional grade tooling i realized that it's a work in progress and that it's not the immediate priority of the space but it will make a lot of institutional capital more comfortable with the space as a final shameless plug we're incubating fractal protocol the attempt here is to build a fixed yield protocol that will abstract all of the above everything i've described from the user today we will allow you to deposit from any chain source a fixed yield we we in the back end we implement various yield generating strategies and um sourcing the yields from across chains the thesis here is that really two main points if you believe that d5 yields are going to be an order of magnitude greater than anything else in traditional finance for the foreseeable future there needs to be a simple on-ramp and secondly that chains eventually need to be a back-end system that gets abstracted away from the user thank you very much [Applause]