Episode 370

Diving Deep Into Liquidity Mining

Kain Warwick
  • Play

  • Play

  • Play

  • Play

  • Play

  • Play

  • Play

Crypto is largely reliant on incentive mechanisms, and liquidity mining is one of the more recent cryptoeconomic incentive models to emerge. When decentralized exchanges need liquidity, they can leverage liquidity mining to incentivize users to provide it. In turn, the “miners” generate revenue, generally in the form of a native tokens, proportional to their share of liquidity in a pool. This summer saw a surge in activity surrounding this concept, with Synthetix a notable player.

Kain Warwick is the founder of Synthetix, a company creating synthetic assets for DeFi, enabling exposure to fiat currencies, commodities, and cryptocurrencies. Hasu is a crypto researcher and writer with a focus on game theory and economics. They joined us to explain the key concepts behind liquidity mining, how and why it was created, and its increasingly important role in DeFi.

Friederike: Hi, this is Epicenter and I’m Friederike Ernst (and I’m Sunny Aggarwal). We’re having a bit of a special episode in that it’s not on any one particular project. It’s on liquidity farming in general. We have two guests today, Kane from Synthetix, who we’ve had on earlier this year as well. Hasu maybe both of you guys can give a very brief introduction on your backgrounds and who you are. Kane, do you want to start?

Kane: Yeah, sure. I’m the founder of Synthetix, which is a synthetic asset platform built on Ethereum. We allow people to get exposure to a range of different assets like gold, silver oil, et cetera.

Hasu: Yeah, Hasu. That’s a pseudonym, not my real name in case you wonder. Before crypto, I used to play online poker professionally for 10 years, played pretty high stakes for a long time. Something that I increasingly learned was that I enjoyed the studying aspect of the game more than playing and basically discovering the hidden roots and strategies of the game and then teaching that to others.

When I was done with poker, I looked for a space where I can apply that same approach basically. I got very quickly sucked into crypto and I would assume you would all agree that crypto is like, you have a million strings around you and you can pull at any one of them for years and basically you never get to an end. I think that’s so fascinating about it.

I basically do research full time, but I just do it for myself, invest my own money and for the last year or so I also run the research desk for Derebit, the largest options exchange in the space. We write weekly research columns, that’s me and Suzu of Three Arrows and a few months ago we also started to podcast together called uncommon core. That’s what I do pretty much.

Friederike: It’s super interesting. We are linked to it in the show notes. If you guys need more podcasts, I would totally recommend it. Let’s talk about liquidity mining. The entrance of liquidity mining has been somewhat gradual. If you look at the earliest instances of liquidity mining, when do you think you would say it actually started?

Kane: There’s been a few different iterations. Certainly things like Live Peer, for example, were early iterations of trying to work out how to do some mining distribution that wasn’t at a base layer. Something above the base layer that allowed you to distribute tokens or assets based on some provable piece of work that you’re doing.

Hasu: Yeah, I agree. One thing, when I first heard about liquidity mining and that was actually not when Synthetix did it too. That was probably the moment when most people first heard about it, but when compound did it this year and it basically hit the mainstream in a big way. When I heard about liquidity mining, I immediately thought, okay, this is basically proof of work mining. This is basically mining as we within Bitcoin aren’t as we haven’t even in proof of stake at cosmos, but in a more generalized way. In these networks, like Bitcoin or even the dApps, they also say networks, if you want to think about them this way, you have different groups of actors and often one group incentivizes the other to perform certain work for them.

Hasu: Yeah, that’s basically what we have in Bitcoin right? So the queen, you have the users who want to transact with each other, want to find out about their transactions and the way we get this, the users pay mine, us to collect transactions in the block and then test them with a cryptographic proof of work. That proof of work is what creates finality over time cost basically makes the transaction is Three harder and more costly to reverse.

I like that you brought up Live Pee. I wanted to go into that direction in terms of explaining. It all started with Bitcoin, but then we already had back in the day, different basically the network incentivizing different things that were not just attestation of messages, for example, we’ve had storage. We’ve had stuff like, I don’t know how to pronounce it, storage, and then SIA, for example.

That’s where actors in the system are paid to basically provide store data and then provide proofs that the data exists and so on. The way I think about liquidity mining is basically a generalization of that concept as it relates to providing financial liquidity.

Sunny: In Storage and SIA, it was mostly other users paying other users for storing data. Would that still count in this liquidity mining or does liquidity mining necessarily imply that it’s some sort of new supply that’s being distributed?

Kane: I think for me it needs to be a protocol level incentive. That’s designed to bootstrap before people are willing to pay Or augmenting what people are willing to pay. Maybe that’s not strictly necessary. You could launch networks that immediately had sufficient network effects that people were paying each other enough. At that point I would say you almost don’t need liquidity mining. You just have payments. I do think there is a component of it that’s about generating those network effects and bootstrapping.

Hasu: Yeah. I think that’s a good point. On the one end, the Bitcoin block subsidy would go away. We wouldn’t say Bitcoin no longer has mining. It still has that aspect. The way that we talk about liquidity mining, the way that it’s used in the public discourse is basically synonymous to a subsidy that’s basically supposed to create network effects early on.

Friederike: When I think about liquidity mining, there’s actually, obviously network effects and basically I would call that growth marketing in the broadest sense. Obviously they play an important role, but there’s also different aims here. Right? So basically there’s programmatic decentralization to a certain extent. You want a broad distribution of your token and an inclusive governance and you want, I mean, basically the entire fair launch movement grew out of this.

You also ideally want closer alignment between your own token holders and the people who use your protocol. Right? Because, if you look at the token distribution after the 2017 ICO bubble, a lot of it was just for speculative purposes, despite the fact that almost all of these tokens were branded as utility tokens, but people held them for speculative purposes. Is the core aim for you guys of liquidity mining? Is that growth marketing or does it have a deeper ideological?

Hasu: I think it has two very practical goals. You basically touched on both of them, the one that’s always to incentivize the provision of some work or service to the network and you need that early on in two-sided marketplaces because, for example, on a compound, you won’t get any borrowers unless you have lenders and vice versa, then all lenders and borrowers.

That’s why it’s important early on to bootstrap or to provide an additional subsidy beyond just the supply or the utility that you can already get. The second is the initial distribution of supply. You have this fair launch theme. Coins that had a fair launch everyone had to pay to acquire the supply and everyone had to pay the same market. That’s the definition, not no one got any coins for free. There’s no pre mine, no VC allocation. That’s basically that’s the fair launch.

We’ve seen that again and again, that coins that do have a fair launch get assigned a premium by the market. Those coins are seen as more valuable, the distribution is seen as more fair and that’s why more projects do it. You can also see liquidity mining basically that different projects use liquidity mining in very different ways. If you look at Yearn, for example, Yearn also used liquidity mining, but the work they incentivize was just to lock up liquidity in a pool that no one can interact with.

It’s a staking, but the staking doesn’t serve anyone. It’s completely useless work or that maybe that people go to the website and that it builds brand awareness and so on. That’s an example of a project that completely focused on the second part of liquidity mining, which is to distribute all of the initial supply into the hands of the market. That’s what Yearn did. I think it was two to three weeks. Then a hundred percent of all YNY was already issued.

Friederike: I would actually be interested in Kane’s point on this, because I mean, at Synthetix you actually saw a fatty early on liquidity mining program. Basically what did you guys hope to achieve with it at the time?

Kane: I think that look there’s a lot to unpack there. I think that we can go into fair launch and what that means and the ideology behind it. I think there’s a practical aspect to it. There’s correlation between fair launches and very militant, devoted communities. The causation is actually the wealth effect. It’s the fact that people who mined Bitcoin in the early days, their cost basis is zero. The people who liquidity mined wifi, their cost basis is zero. If you can get someone’s cost basis down to zero everything’s upside for them.

If you look at the projects that survived through the bear market in 2018 at one point the price of a SNX was down to 2.50 cents. There were people who were buying it for every seller, there’s a buyer. Those people are deeply devoted. Their livelihoods, their identities are very strongly tied to those things.

I think that enables you to distribute tokens with effectively zero cost basis. There’s an opportunity cost, obviously but I think that alignment then people realize that there is this there’s an ideology and there’s a fair component to it. I think there’s also just a pure economic incentive component.

Sunny: Could you explain a bit about how did the liquidity mining process work in Synthetix, what exactly in the Synthetix system was being incentivized and how did the process work?

Kane: Yeah, so we needed, we had this magical walled garden. If you were inside of it, you had this cool property where you could take any asset and go from one asset to another with zero slippage, but you had to get in there first. The problem was that the walls were way too high and there were no doors and it was like, you literally couldn’t get into the thing.

We had this idea within our discord to incentivize the lowest risk pool that we could in Uniswap, which was like the synthetic Ether. There’s very little risk for people in terms of loss. We put incentives in there to essentially allow people to go in and out of that pool. That was the most liquid honor, but all of a sudden we had a door and people could go into this magical place. That was really the initial incentive. Then it grew from there, we added different incentives and different structures.

Sunny: It was incentivizing people at the on ramp into the Synthetix system to allow people to easily acquire the SNX token.

Kane: The actual synthetic assets themselves. You wanted to be able to convert a synthetic dollar to a synthetic Euro, for example. You needed to get one of them first to start that process going. We said people have Ether, we’ve got a Synthetic ETH we’ll create a pool that lets people go from ETH and dedicate, and then you can make the next hop to any other asset.

Sunny: I see. I see. Why do it with synthetic Ether instead of with SNX, which then could be converted into any of the synthetic assets.

Kane: There were debates about like, should we incentivize SNX liquidity? The real reason was basically impermanent loss risk. The only people that were going to be able to do it were fairly sophisticated about the risks of Uniswap; we didn’t have the masses coming in and providing liquidity and Uniswap at that point. There was a lot of skepticism around, like, who’s going to step up and put SNX and Ether for example in the Uniswap and take on that aisle risk

Hasu: 2019 was a different time. In terms of what coins had the most liquidity, because David Cohen said below $5 billion, or even way less than that, of liquidity. Nowadays it’s like 20 or 30. Nowadays incentivize the SUSD versus USD ports and not Ether versus SEther for example.

Kane: Absolutely. We did one Curve came out and we had the sUSD, DAI, USDT pool that was incentivized.

Hasu: Yeah. I think that’s, Synthetix, this is really the exact opposite of Yearn of the example that we gave. Basically the subsidy said zero purpose. Other than distributing supply and Synthetix is very picky about distributing supply. You guys do basically trial runs where you incentivize stuff for two weeks at a time to see how the different agents in the system react to these incentives and it then doesn’t react. Then you basically change the parameters or try something else. That’s, I like that a lot. Yeah.

Sunny: When we say liquidity mining, does it have to be something useful that’s being incentivized. Would we call the Yearn incentivization or just this idea of like generalized staking, like when I staked all my tokens in the Yearn protocol, was that liquidity mining or is that something different? Then even when we talk about liquidity mining, does it have to be literally liquidity that we’re incentivizing or is something like Live Peer called liquidity mining?

Kane: Yeah. I mean, that’s why I think yield farming ironically took off. Because farming doesn’t really mean anything. There’s a yield in this governance token or protocol token that you’re getting for farming at which could be literally anything. I think that’s why a lot of people it resonated with. I remember when it was first coined for me, it was like, Oh yeah, that makes sense. Like, this is not presuming anything. It’s just yield farming. It’s very adaptable.

Hasu: There wouldn’t really be much point for us to come up with any definitions, because we will have a hard time convincing the rest of the community to adopt them. It’s just liquidity mining, like mining, generalized mining, liquidity mining, I use synonymously. If I had to give an attempt and I think mining is the act of providing cryptographic leaders that verifiably work for another party in the network and that’s incentivized by the protocol.

Then it’s basically liquidity mining. When the work that you do is providing liquidity and it’s yield farming when it’s, that’s basically what the user called and said, when it feels like putting money in the bank account and then getting some Ute for it. That’s I haven’t really seen it to you as an in when the work is not just providing liquidity.

Sunny: Okay. Can you talk about your experience with Synthetix? So basically the people who provide liquidity and the incentivize pools, do they typically also become true users in the sense that they interact with the protocol when it’s not incentivized?

Kane: I think so. The interesting thing with the initial yield farming incentives that we had is we weren’t directly incentivizing you to interact with the Synthetix protocol. You had to get synthetic ETH. You had to have it from somewhere, but you could have just bought it on Uniswap. We were actually incentivizing people to use Uniswap. Which is funny when retroactive Uniswap rewards where dropped on all these early SNX people’s added hundreds of thousands of dollars worth of uni for just being around, and maybe putting a few thousand dollars with liquidity in there. I think it was initially incentivizing people to use Uniswap, but we got a lot of utility out of people using Uniswap and being LPs on Uniswap that was our OnRamp. We thought we better use this thing. It’s going to be huge. Let’s ride on the coattails of that rather than trying to build our own liquid on ramp and, or build our own DEX or whatever. It was about leveraging what was there.

Hasu: It’s interesting that this watershed moment for liquidity mining actually happened before liquidity mining went mainstream. In defy, you can incentivize things that happen outside of your own protocol, because you can observe that this thing happened and it’s directly provable. Different base layer blockchain have a lot of trouble verifying stuff that happened on another blockchain.

That’s the whole story behind the interoperability phenomenon. That’s what I think is the most interesting thing about DeFi that you have this one highly composable database where everyone can interact with what everyone is doing. Liquidity mining is really as experimenting with programming, humans in this composable environment. It’s extremely fascinating.

Sunny: Yeah. I think one of the things that became really interesting was the fact that all the data on the chain is readable. Now you, ability to access any sort of data and incentivize it in a way that maybe wasn’t possible before. To what extent do you think like AMMS in this liquidity mining? I was pretty excited about this project called humming bot. That seems like it never really took off, but it was basically incentivizing market-making on order book based dexes, but it never really quite took off. Why do you think that was?

Kane: You have to run your own infrastructure? Which just immediately filters out 99.9% of people. Before we jumped on here, I was unplugging and plugging in my dApp node, trying to get my ETH2 Two validators up and running and not having the right packages and stuff. If it wasn’t ETH, there’s zero chance that I’m bothered to do that. I just wouldn’t be.

I think it was too difficult versus something like Uniswap where there was a learning curve. No question, especially the early UX, but it’s maybe four-five clicks and all of a sudden you’re a market maker, maybe not a very smart market maker, but you’re a market maker and you’re there. I just think that made it so easy for people who had these assets that were just sitting there and wallets doing nothing to go well, maybe I should I should put them into this pool and get some utility out of them and then potentially some of these.

Hasu: Yeah. For the other side as well. That’s the very point in providing liquidity. Something like Uniswap when it doesn’t generate fees and the only people trading against it are the arbitrageurs and then you bleed out from the divergence of those. Uniswap and DeFi really benefited from this peer to peer, with moderate or peer to contractors. It’s also sometimes quite where something like Uniswap always will trade against you when you need it.

That’s the building block that a lot of other this invisible building block, that’s actually the foundation of pretty much everything that’s happening in DeFi because every project is in some shape or form depends on being able to liquidate collateral of users for example.

They rely on Uniswap in order to get that liquidity when they need it and to be able to liquidate the customer’s positions. When you don’t have that assurance that in any market condition, there is liquidity there that someone will quote you basically liquidity, then these projects wouldn’t work. I think that’s basically the invisible foundation for all of defy.

Friederike: Shall we talk about the elephant in the room? That being impermanent loss, anytime your liquidity provider is an automated market maker as a liquidity provider, under what circumstances should I actually provide liquidity, given that the underlying assets are volatile? That’s why you guys started with sETH against Ether. What do you think the rational approach is here?

Kane: I mean, it really depends on the intent of the person. There’s certain pools where if you’ve got a lot of one asset. This is where balancer, I think, helps. It makes us a little more palatable. It just opens up the number of people that can rationally provide liquidity. Or less irrational and provide liquidity, If you own a lot of an asset and you’re willing to sell that asset down continuously, for example. You’re not going to just sit there and hold it at some point in the future, you can liquidate all of it at a better price. You’re going to get a continuous price from whatever the ratio is now to whatever the ratio is in the future. If you have a long time horizon and are comfortable holding both those assets.

Let’s say you’re holding Bal and Eth. I’m a Bal\Eth liquidity provider in balance. I’m very comfortable that balancer is an asset that I’m happy to hold, but I’m also happy to sell some down at the process running up, and obviously I’m happy to hold Ether. I think under, circumstances, it’s totally fine to be in there and be in that pool, as long as you understand what the trade-offs are. If you’re someone who’s saying I’m buying, I dunno, Ether today at 500 and I’m planning to sell it at a thousand and you want to wait to left targets hit, and then like you shouldn’t be providing liquidity in Uniswap like, that’s not going to get you that outcome. I think specific strategies that you want to be employed. I think for the average person, it’s I’ve got this thing, dump it in what happens. They don’t really think too much about it.

Hasu: Yeah. I don’t really have a ton to add to that. I think in general, people should be aware of what they’re buying when they’re buying pure tokens. Structurally when you put money into a liquidity pool and what we basically sell the two assets that you put in and you buy a token, that has very different behavior than just sorting the two toppings initially. I thought there was a small chance that this would lead to disaster, basically at the height of the last DeFi run up when first sushi swap and then Uniswap, and then all the copycats started, basically everyone to become an LP, even though had never interacted with them before. They, these people mostly had no idea what financial product they’re buying.

They were actually selling liquidity basically, and nobody really told them that. They were attracted by the allure of the liquidity rewards. The rewards turned out to be so big because the demand for these liquidity tokens on the market was so high that no one could have really lost any money there. I guess now a lot of people know how liquidity provision works and what they’re actually buying, it was a good education in that sense. In general, I think it’s good if people want to engage with this, know that they’re selling liquidity and hence look at the correlation between the two assets that they’re putting in and the more correlated they are to each other on a longer timeframe, the better.

That’s why you see that actually in the liquidity mining rewards, which is interesting because that’s basically the market telling you what is the cost of putting assets in there. If you want to get a feeling for that, just look at the reward of different Uniswap puts and different balancer pools pay you. The more you can earn the more vested the reward, the market demands in order to be in that pool. I thought that’s a nice way to put it.

Friederike: Yeah. That’s a nice way of putting it. Can I loop this back into the big economy? So basically you said liquidity mining was so profitable because that basically no one could have possibly lost any money. This is very counterintuitive coming from a traditional point of economics. Where did this money actually come from?

Hasu: I should clarify what I meant with that

Kane: Yeah. I definitely lost money. I will say that 100% not every time obviously, but there were some cases. Based was a really great example. I managed to somehow buy the absolute top. I was liquidity mining based, I’d bought at like 200, put it into a pool, and I was like, Oh, this is going really well. I’ll just double down 10X down actually is more accurate probably. Bought a bunch of based at 600 and then got utterly crushed down to 150 and then capitulated at the bottom. It was a pretty painful lesson in a permanent loss. Like, so there were all the things obviously that I did pretty well in, but there were some examples where if you like, even, I would say somewhat knowing what you’re doing, you could still get your face ripped off pretty effectively.

Hasu: Oh yeah. I should clarify that. As Kane said that there were different pools that tended to have different labels. The pool one is basically, let’s say you provide Ether B2C liquidity, and then you get paid rewards for that. It was quite hard to lose on that. Cause a divergence loss relative to fees is very small. If at all, you mostly wouldn’t without liquidity rewards even.

Then there was usually a second pillar, quite the poor Two and there you basically, you provided the projects native talking, let’s say based against some of the money, let’s say Ether or USB-C. That was basically to provide exit liquidity for the people get entry and exit liquidity for those gambling on the project. The rewards on those were in the thousands of percent per year.

It was also incredibly easy to get wrecked in those pools, because it was basically a game of chicken, basically one day on the large short we wanted to exit, they could do it through this pool. Then if we would immediately collapse because all the piece would stay in there, they would end up with all of these native tokens and they would sell off all of their Ether. Yeah, it was incredibly easy to get wrecked there. Many people did for sure.

Friederike: Okay. They were definitely pools where making money was not a given, but the larger lot of people got very rich with yield farming. Typically if you don’t think at the scale of nation’s, money doesn’t just appear out of nowhere. What happened here? Is it a zero sum game or not?

Sunny: If not, then where is the excess value coming from?

Hasu: It comes from the people buying the liquidity, talking on the market. I mean, look at curve for example. You get liquidity rewards in the form of CRV, tokens and curve. If you provide liquidity to the pools and basically the how high the reward is for you depends almost entirely on how much you can sell the CRV for on the market. The meme that the retail CRV bias was the foundation of the entire you’d farming hype. Because someone has to buy these relatively useless tokens at the time anyway, and CRV for a time at a fully diluted market cap than Bitcoin even That was quite funny. It has collapsed predictably, I think over 90% since then. The money, created from thin air in that sense, it’s coming from the people who buy these liquidity tokens.

Friederike: Let me dig in there. Basically the liquidity, I mean, basically the tokens that are, for instance, in the case of CAF, it’s a governance token and most of these tokens are up. Would it be correct to say that in a way you’re buying future equity or future to become a shareholder or in traditional terms that shareholder off that project by buying cuff tokens retain, are you in effect buying a bit of cover as a project? So basically in a way to actually incentivize with future bits of the project.

Kane: Yes. The problem with curve is you’re the vision fund. Buying it at a $48 billion valuation when the market doesn’t really believe it. I think that, is there’s two sources. One is reflectivity. These are highly reflexive assets that didn’t exist. People are holding them and we all believe collectively that there’s some value to curve or Uniswap, or any one of the new crop of assets, like yam, et cetera. That popped out of nowhere.

As long as there’s enough of them and they’re sufficiently distributed and sufficiently illiquid that some people are gonna hold on to them and feel like they have value. That’s where a big portion of this new money came from. Then I think on top of that, you then had the fact that there was energy coming into the system from Bitcoin flowing onto ETH for the first time, all of a sudden this BTC, just crazily flowing from somewhere.

It wasn’t doing anything sitting on my cracking account. I’ll tell you that much. It’s now old, my BTC on Ethereum. All of that BTC flowing in and buying things like that. Purchasing power turned up. I think it was a combination of a few of these different things that created this new value. That there previously wasn’t. It was technically there. Because curve existed pre-token and had some value, it was generating fees. All of a sudden those now just this tangible liquid ish representation of what curve was That was tradable and you then multiplied that by 20, 30 different projects and you had stuff there all of a sudden.

Hasu: Yeah. Yeah. I would agree. I would also agree with you Friederike that’s. I mean, these it’s definitely To think about these governance talks, even though they’re quite governance tokens as a form of pseudo equity and what it can look like. I mean, the current bias who are buying an XCRV and these other talkings at seemingly very high prices, if CARF would end up replacing the large Forex exchanges or whatever, then of course, that bet can also pay off. It’s a sort of venture bet. I think that’s the right way to think about it. If it succeeds, then you would have incumbent financial service providers lose and a currency of liquidity providers win.

Sunny: We mentioned the idea that AMMS were important for the rise of liquidity mining. A couple of weeks ago, we had SPF on the show and there, his claim was that like a lot of the incentive was the other way as well, that AMMS only took off because there was this giant amount of liquidity mining that was incentivizing people to provide liquidity and that they probably wouldn’t have otherwise, how valid do you see this claim?

Kane: It’s not a very credible claim. The data doesn’t support. I guess it depends on what your metric is, how much volume, how much liquidity, but I mean, even curve. There were no incentives for a curve, you know? There were two, 300 million in liquidity that was in curve because it was generating fees. It was the best yielding thing pre the crazy yield farming, it was the best yielding location for your stable points. It was the only yielding location for your sample ones. It was the only place where you could put your stable coins in fairly comfortably brisk adjusted. The yield was amazing, you know? That’s been washed out with these massive returns from seniority, but like, that was a good thing. It made sense and it was tangible. It was real.

Hasu: Yeah. The week before sushi swap even launched, which was the first big scare liquidity subsidy for four AMMS, Uniswap made headlines in mainstream media because it actually surpassed Coinbase and transaction volume. Then before any liquidity subsidies you have to make it even more liquid, even in some past where the traditional exchanges were dominant up until that point. I think the rise of AMMS was unstoppable even before that.

Second, I think it’s also an unfair comparison because you really get to, if you start a traditional exchange like FTX, then you raise venture capital typically. Or what they did is they’re sorta talking and raised hundreds of millions of dollars from the market. Then that float back into their market making operations.

Starting a traditional exchange is very difficult. It has the same problems of bootstrapping liquidity. You won’t get any stakers unless they’re makers, so they all start on either by making markets on their own exchange, and a lot of exchanges get super wrecked making markets on their own exchange because market-making is very difficult. When you’re making markets in a poor way, on your own exchange, what happens is you attract the smart takers, we’ll pick you off whenever you’re quoting some wrong price and you will bleed out hundreds of millions of dollars over the first years of your operation.

What they tend to do is they take the money that they raised and they incentivize basically market-making by professional market-making firms and they also pay, so this is just the same. This is just, they pay liquidity rewards for market makers. It would be really unfair to say what AMM only took off because of liquidity mining, because they use subsidies to check market-makers wire, audible based exchanges, do the exact same thing, and nobody talks about it. Right? Either we accepted in both or, I mean, you would see none of these large exchanges, if they would give, maker rebates

Friederike: that’s a very good point. How sticky you think the Liquidity provision is sushi swap for the liquidity from Uniswap and then they swapped it back and then Uniswap upgrade to V2. Basically, a lot of the liquidity migrated incredibly quickly. What’s the sticky part? Is it the brand Uniswap or is it, I mean, it’s, can’t be the contract?

Kane: I think if you take a step back, right, the stickiness is in the market itself, regardless of which contract it’s in, there are people with assets willing to provide liquidity, regardless of whether it’s in sushi swap or curve or Uniswap or whatever. I think when people look at it and go, Oh, well, it went from here to here and it’s bouncing around, like, those people are optimizing for the best yield they can get at that time. They have bought into the idea that providing liquidity, being an LP in an AMM is a thing. Some AMM is going to exist. It’s created this network effect, which I don’t think is going anywhere.

Hasu: Yeah. I think this touches on a very important question, which is what are we actually incentivizing? This is what Synthetix is so good at in my view because they only incentivize liquidity when they have a thesis that they want to validate and they know exactly what they want to test.

When you create these large scale incentives for market makers in Uniswap, for example sushi swap doesn’t seem as focused. In my opinion, they should incentivize to own the customer, a focus on owning the customer. I think that is the thing that is actually sticky, that they use us who go to the Uniswap app and to then take basically their business there, because we know that for the LPs themselves, they can switch with the click of a button.

It’s the users who are more sticky and it’s not sure yet that a system like Uniswap or like sushi swap that they can own any customers long-term or if basically the customer will be entirely on by aggregators like one inch or Matt Shaw who basically just searched the market for the best route. Then the users order through there. Then they are the ones charging them, then all the fees on top of it.

If it goes even a step further than that, if it would be MetaMask or Argent wallet or Coinbase wallet, who would end up owning the customer. We have no idea yet at this part of the market cycle, in my opinion. It could be that all of this liquidity incentivization would be for not in the end, in, I think that’s entirely possible.

Sunny: Kane, at the beginning of the episode, you mentioned that like liquidity mining is most useful when you’re trying to create a network effect. How do you explain this uni token, where it’s like Uniswap clearly within the DeFi space has already such a strong network effect. What was the point of introducing liquidity mining there?

Kane: It was pretty obviously a reaction. I think there was an intent at some point to deploy a token and I’m maybe the most out there person that I know of and have been for a long time in terms of token maximalism. I believe that tokens are inherently valuable and I think the design space of tokens and using them as the coordination mechanisms were not even close to having explored that. I think that there’s so much more that we can do around coordinating behavior.

When we look at Uniswap and say was it that effective? Like, is it wasting? Probably honestly, if I have to be totally honest, it is. It’s also distributing the token and someone needs to decide whether they want to hold it or not, and therefore what they will, what they will do. Like, there’s some long-term alignment in terms of people who are holding uni tokens.

Friederike: I mean, what you needed, interestingly, that a lot of the other liquidity mining programs didn’t do is they did this backward distribution. Basically anyone who was a regular and possibly even privacy conscious Uniswap user over the past year or so got a lot of unique tokens. Don’t you think that the way that Uniswap actually did the distribution targeted an organic distribution of the uni tokens among the users more than the forward usage?

Kane: I mean, I think it was a second order effect. I don’t know if it was the primary goal to do that or not. What I will say though, is that it did what I had described earlier. Which was established for a ton of people, a zero cost base on their uni holding. Of course, some people, if you give them something for free, they dump it. Of course. But the market was pretty happy to absorb that selling and we found an equilibrium fairly quickly.

We’re now in a position where there’s a lot of people who are still holding those uni tokens that are very bought into the project. The issue, I think that Uniswap has is because the token is so new, they haven’t had a chance to get that community in the way that some of the other projects out there have. Like an Ave, or there’s a bunch of examples. Where the token has been around long enough that a community is formed.

There’s a lot of uniswap users out there, but if you go into the, I mean, I remember being in the Uniswap discord and just being like, this is fucking incomprehensible, what these guys are talking about. They’re a bunch of engineers in there and it was very smart people talking about very hard problems. I thought it was cool, but it was way above my pay grade.

I don’t think that there were that many people in there that were talking about usability or excited about the project, from a user perspective. I don’t think that communities ever really emerged, even from what I’ve seen today. At some point that needs to happen if Uniswap is going to have a user driven community driven brand awareness and growth mechanism. I just don’t think it has that now it’s getting away without it, but it would be far more beneficial if it had it.

Hasu: Yeah. That’s one of the main benefits of a token. In my opinion, that it crystallizes community this way less just incentive to engage in a community that doesn’t have a token. Because you’re not financially invested in it. Whereas with projects that do have a token, we have this double incentive to be part of that community. We could talk about the retractive distribution of Uniswap.

I think that goes back to liquidity mining, which is one, distributing the initial supply. Second is to basically incentivize future behavior. If you retroactively distribute talkings, what’s that gonna do for behavior of people in the future? At least not in a very direct way, but I mean, the second aspect is also important. If you think about what happens every time when you say, okay, the people with the biggest pockets get out of the toppings in a highly predictable way, then you always get the same outcome.

It’s the same three to four firms that, or 90% of your token was, we’ve seen this with many projects. I mean, most people don’t really notice it. It’s true. That’s why projects are looking for ways to distribute some talkings also in a way that it’s not Web Two, but just the amount of capital that you have is not the only factor that is looked at, but in effect does so hard to do in a non-game of a way. The only way I like the way that Uniswap did it, which was way very elegant is to look at something that has already happened because then it cannot be gamed anymore and I’m giving it away to these early adopters I thought was a great way that I would think that some of the projects would probably copy in some small way, at least.

Friederike: In a way Uniswap actually did a 180. They were always this community darling. Then I know a lot of people who personally would have loved to invest into Uniswap, actually went the VC route and then after the entire sushi thing happened and then liquidity was forked, they came up with this uni token and gave it out to the community. Do you think that was intended or at least that’s my stance it probably wasn’t intended from the get-go. Do you think this will put an end to these VC offerings of fairly established projects in this space? Do you think, I mean, Kane, we all know that you’re a token maximalist, but do you think, I mean, Uniswap could have done a token earlier. Instead of going to VCs, do you think we would see that more and more?

Kane: I think the reason why they didn’t do a token is not because it was a choice between VCs and selling a token. I think it was anti token rhetoric that had infected the core Ethereum community of which Hayden was paying a lot of attention to. There were a lot of people in the community that had a very deep distaste for what happened in 2017. Even all the way going back to 2016. This idea that if you had created a token ever read the ERC 20 contract, whatever, like if you’d even come within a hundred feet of it, that there was some black mark against your name. That was deeply embedded in the community for a while, which was this just very visceral reaction to the craziness of the ICO boom.

I think it over-corrected massively. I think had that not happened, we probably would have been much more open to doing a token. Maybe it would have been like VCs and a token in parallel or something like that. It was a very clear and obvious anti token stance from a lot of projects at the time. I don’t think Hayden did this. A lot of people would come out and even use it as like, Oh, we didn’t do a token. Go to their website. The first fucking line is like, we didn’t sell a token. We never did an ICO. That was like a rallying cry, for a long time in the community. I just think that was misguided. I think it just, it was not really beneficial in the end. It was an overcorrection.

Hasu: I think that’s kind of, you have a market cycle within the market cycle in crypto and acceptance for tokens is definitely one of them. Everything needed to token even stuff that really doesn’t need a token 2016, 17. Now I feel like we are maybe a bit over-correcting in the other direction right now. Right now, everything should have a talking, even if it maybe doesn’t make sense, you see this with governance. Governance is not just, I mean, governance is used as an excuse right now to launch tokens when the governance is really not something good for the project, but it’s more often the tech vector than actually is maybe necessary for the project to work. Something like compound, you could argue that there needs to be someone who sets the safe collateral limits who can add new toppings, approve them and Two, and the interest curve and whatnot.

Kane: That should be to ruin just sitting in his ivory tower, just tweaking.

Hasu: Yeah, exactly. I think that’s where they’re going, parameterized tuning of as a service and then get paid and talking to self fees. Very interesting business model, but yeah, I mean, you can have other projects, like, I mean, for, Uniswap definitely, you can argue that the token has a lot of benefits to them in terms of crystallizing community, but what is really the benefit beyond that?

Luckily we haven’t really seen any big takedown of a project with governance token, like they Roque approved some dramatic change and still have customer funds. I mean, we could have seen it with maker where anyone could have, I think it was 4% of the total MKR supply that would’ve been enough to stake them on a new executive contract and that contract could have paid a billion dollars to have the heck out.

Now a bit later we saw that the MakerDAO approval was actually a make-up proposal that was approved with a flash loan. We saw the first interaction between those two concepts. I think we will get to a point where the community starts to see a governance token assessed as a risk vector again. Then we might see a correction another direction, again, in my opinion,

Sunny: What are some projects right now that you think would most benefit from some sort of liquidity mining system that maybe don’t have it now?

Kane: Tornado catch? I mean, yeah. Anyway, that’s a long story. We’ve been talking about that with the two Romans for a long time. I think it will get one. They maybe in the early days fell into that same anti token mindset a little bit and I don’t think it needed, you don’t need a token to build a project clearly. You don’t even necessarily need a token to govern a project.

Even with us. Our recent governance moves, the fact that we’re not even using the token as the waiting for governance anymore, we’re actually using your percentage of the debt pool that you represent. Which doesn’t even really necessitate a token at all. We could have another form of collateral and then be weighting it against the debt pool. This goes back to the point that Sunny made about you can read anything, we could use your Uniswap holding is our governance if we want it to.

If you’re a union self LP, you can come and governance and that’s it, there’s nothing stopping us from doing any arbitrary read within the Ethereum state and then using that as governance. I think you can still govern a thing. It’s just a question of like, is the incentive alignment going to be right there if we’re using Uniswap LPs to governance, Synthetix like, is there any misalignment there? I think in some cases, yes, in some cases, maybe no.

Hasu: I give another answer. What project would benefit most from liquidity mining? I would say the lightning network for Bitcoin and really any layer 2. Do you have the same problem on Ethereum? You won’t have this problem with optimism and other chances, any layer to the basically blockchain of its own and starts with your network effect. I think you will see liquidity mining there to basically bootstrap the network effect. I’m way more optimistic that optimism will succeed with the lightning network, because of this aversion against tokens. That’s very unfortunate.

Kane: I just went down like a micro rabbit hole there for a second. Like, I feel if you did try to propose some protocol incentives to lightning network, A: it would be contentious, everything’s contentious, it’s Bitcoin. B: it would end up with some contentious hard fork. Where one version actually incentivized lightning network and the other one big block, small block an internalizing battle. I think maybe ironically, the lightning network incentivize a version of Bitcoin that pays out like BTC level block rewards to the Lightning Network could actually be more successful.

Hasu: I’m not sure how it would lead to a fork. The interesting thing is you could do it in a way that doesn’t break the Latin at work. I mean, if you would have a second form of channel that pays out future equity in its routing fees to people who route money through it today, or whatever, I mean that could exist in theory. Anyone could interact with it without partaking in liquidity mining itself. I mean, but the reason that we don’t see this is of course, because of the huge social barrier against entertaining any form of token in the Bitcoin communities. I don’t think it really it’s the technical hardest. It’s even the threat of a hard fork, purely that the social pressure on anyone entertaining these ideas.

Sunny: It is possible to have the source of the liquidity mining incentives be from something other than the base protocol or the lead development team, not for lightning network. I was once a couple of months ago, considering trying to build something for Interledger to do exactly this. How do we create incentives for people to participate in the Interledger network?

Even before the uni token was launched, there was this short-lived project called uni Dao, which was like, it was just this reaction to the sushi swap where they were like, let’s airdrop tokens to all the Uniswap LPs. They asked me to be a multisig signer for that. I’m like, sure, it wasn’t too much involved otherwise. So far we, seen one of these really takeoff where all the liquidity mining has always come from the protocol or from the lead development team. Do you think we will see some successful ones from other sources? Eventually?

Hasu: It’s a very interesting question for sure. I mean, in theory you have this huge developer funding system on Bitcoin that’s very mature. You have a lot of grant-giving institutions and in theory, nothing prevents these same institutions to incentivize liquidity on the Lightning network. If they think that’s the best thing to make Bitcoin grow, you don’t necessarily have to pay developers. You can also pay liquidity providers. I think that’s totally viable. I don’t think we’d see it in Bitcoin, but in general, I could imagine, for example, the Ethereum foundation or any other foundation could also incentivize these programs.

Sunny: If you think about it as sushi swap for a while, it was basically incentivizing liquidity on Uniswap.

Kane: Yeah. I think when there’s a power vacuum. When there’s no clear intent to step in that’s when the opportunity arises for someone to say it’s not going to coordinate itself, we’ll create this new coordination mechanism that maybe exists outside of like, whatever the core team project and go on and do it on their behalf. Right? There’s nothing stopping anyone from doing something like that. In the future, if there’s protocols that aren’t doing watch for that to happen.

Hasu: So far, we have only talked about things that benefit the protocol. You can incentivize liquidity in something like Uniswap to drive more liquidity to it. Then we saw a sushi swap, which was the first instance, okay, let’s create liquidity to take something away from someone else. I mean, this always exists. You’ve probably seen Tara paper saying that proof of stake always competes with yield provided by defy, for example. If defi yeilds are very high, then proof of stake system could have very little money staked, and therefore be very insecure as a base layer.

That shows that liquidity always competes with each other on even global scale. Sushi store was a very direct attack on the under liquidity of Uniswap over. Ultimately nothing would stop, let’s say a nation state to incentivize miners to join the mining pool and that mining pool consensus, basically nonwhite list of transactions or whatever. That is this new era we are in. Where anyone can create the incentive for anything in a permissionless way. We are barely scratching the surface of that. There are some very scary possibilities there.

Sunny: Are there any good examples of liquidity mining sort of being done in pre blockchain systems?

Kane: I mean, I think a lot of the growth that I remember, I don’t know what year it was in the late two thousands reading the Andrew Chin growth hacking ebook, that he released for free. That was mind blowing for me. At that time to read that because some of the examples are so ingenious. Like the example of Airbnb basically tapping into Craigslist, it wasn’t an API at the time, but into their front-end flow, to post listings. They basically, to create this weird overlay on top of Craigslist that would allow anyone to just cross post their Airbnb listing to Craigslist.

Much bigger network effects that actually pulled VAT, Craigslist was getting value out of paying to or from your ad impressions or whatever, paying to view that content, and it was one of the big categories within Craigslist rentals. There was this parasitic entity that… short-term rentals, I think that game was up. I think there are examples where if something is not well defended, people can turn up and siphon off in this case, rental to liquidity. I don’t know exactly what the best term is for it.

I think a number of the other examples were very, not nefarious, but certainly a little bit underhanded, or taking advantage of someone who’s slow moving, not really paying attention, what was going on. They were a startup, they were fast, they had the tech and they built some cool things to just inject themselves into this flow of some other larger lumbering beast, a Craigslist and you’re slacking off their users or attention or money or whatever.

Hasu: Depending on how you define liquidity mining, you can find examples that go back to the beginning of time. I think one thing that basically the blockchain has enabled or has upgraded liquidity mining and that a lot of things that can be now cryptographically verified that the work actually happened that often enables the work itself could be done in a permissionless way and by anyone in the world without having to KYC and so on. That I think is it puts these liquidity incentives into, and globalized them and democratize the stamina.

Sunny: I feel like another piece was oftentimes past liquidity incentives, it’s easier for companies to airdrop dollars to users, but airdropping their own company’s shares is not really feasible in a traditional world.

Hasu: Yeah, totally. It’s frowned upon. Yeah.

Sunny: Do you think there’s gonna be an increase in more traditional companies doing this? Like, I don’t know if you guys saw the snap track announcement where they are giving away a million dollars a year for the top snaps on their platform and stuff

Hasu: On a million dollars to spend on equity, right? Yes. They’re not an owner of the business. Which company was it a few days ago that asked to IPO with security tokens? I’m not sure, but I read some proposals like that. I mean, ultimately I think you would see a lot of equity exists on the blockchain and I think once you were there, then it would be way easier to distribute these shares to users of that blockchain. I think you would see a convergence between the two words.

Friederike: Typically these users are a lot more risk taking than the average human. Do you think the same mechanisms would have worked on average, Joe than did on DeFi Joe?

Kane: I think so. I mean, it really depends, I guess the longer term incentive of owning some aspect of the thing. If you think about a lot of marketplace network effects. Things like Amazon, et cetera, when the third party marketplaces were launched, it was really hard to get people to participate because there was this idea that there was cannibalization. They were just going to cannibalize their own revenue from their direct business or whatever.

I think you could temper that concern significantly by saying, well, you’re going to have if you participate in this new marketplace that we’re setting up, you will have some longer term ownership of it. I feel there’s someone who’s running a business who understands what the potential value of that is versus an end user. It’s maybe a little bit harder.

I don’t think defy users or end users in the sense that we would normally describe them. They’re a little bit more involved in the process. I don’t know the honest answer, whether that will extend to everyone everywhere, but it’s certainly worth trying. Let’s see.

Friederike: You could see it in a way, even as some sort of antitrust action. Because if you look at traditional marketplaces, there’s always a natural monopoly that emerges. If you say, look, there is a monopoly and we accept that, but it’s not going to be a company, it’s going to be a Dao or it’s going to be the users as a group, then obviously having a monopoly is way less concerning than it would otherwise be.

Kane: I think there’s another critical component to it as well, which is barriers to entry. If you, an incumbent that has a pseudo monopoly or whatever, that’s a Dao doesn’t have the ability to coordinate Institute, some sort such that they’re poor positioning in the market is maintained through some extra market activity, Radley that’s outside of just operating and being efficient. Then I think it’s okay.

If the Dao captures the US government and has a bunch of senators or whatever in their pocket, then I think it’s equally bad. We don’t want inefficient markets, even if they’re owned by everyone. I think that’s still a net negative outcome for everyone.

Hasu: I think it’s my theory that we go to, I mean, we’re talking a lot about breaking up monopolies right now and decentralizing the word, but I, that crypto can also have the opposite effect because of how much more difficult it is to capture value in crypto. I think this will lead to there being even fewer, even larger winners who then emerge straight. I think crypto is all about networks and network effects. I think we could see, crypto could have the effect that we replaced three large companies with one huge network. That would be a monopoly of its own then.

Kane: But I think the switching costs are so low. This is, I think, the critical thing provided the switching costs to go from Uniswap to sushi swap. If sushi swap is able to come in and create a slight tweak to the Uniswap model that is much more efficient and everyone can switch very easily and coordinating it is not hard, then I think it’s okay. I think that’s fine. Then you just have this chain of behemoth decentralized entities that is constantly getting replaced. I think that’s actually an okay outcome.

Hasu: Yeah. That’s that would be a fascinating outcome for sure. I think that’s the big promise Of crypto is to lower the exit cost for the user, create more competition by being open-source and forkable, anyone being able to bootstrap incentives basically from thin air. In traditional markets, you have to have access to capital upfront from some venture fund or whatever, in order to even compete with an incumbent business. In crypto, so far, at least you can just create your own talking and say, Hey, this basically stands for any future revenue that we generate and then let price discovery run. Very often these tokens would then be regarded very highly by the market. Barrier to entry for competing in the market and it lowers the barrier to exit for the user, which is in both cases.

Friederike: I think this is actually quite a nice note to end on it’s you know, a good outlook. Kane and Hasu, thank you both for coming on so much. It was fascinating, it’s really fun.



  • 1inch

    Discover the best rates and most efficient swapping routes across leading DEXes. Optimize on gas cost and execute DeFi trades faster with 1inch V2. https://epicenter.rocks/1inch
  • Algorand

    To learn more about Algorand and how it’s unique design makes it easy for developers to build sophisticated applications, visit algorand.com/epicenter
  • cPanel

    cPanel's WordPress Toolkit is the all in one solution that makes hosting your website easier than it's ever been, learn more about it at epicenter.rocks/cpanel

0:00:00 | -:--:--

Subcribe to the podcast

New episodes every Tuesday