DAPPCON 2025: ReCLAMMs- A Passive Approach to Concentrated Liquidity

DappCon 2025 · June 2025 · 18:13

This talk introduces ReCLAMMs, described in its own title as a passive approach to concentrated liquidity. Concentrated liquidity is an automated market maker design that lets liquidity providers set a specific price range for their capital rather than spreading it across all possible prices, and the talk frames ReCLAMMs as pursuing that approach passively rather than through active management.

Speakers

Topics

From DappCon 2025, hosted by Dappcon Berlin.

Transcript

[Applause] Hello everyone. Thanks for having me. Thanks for the introduction. I'm Fabio, data and product at Balancer Labs. And today I'm going to talk about what we have been building at Balancer V3, specifically Reclam, which is our latest innovation on the AMM space. Balancer has always been innovating uh on this space since the 2020. So back in the day we launched V1 with generalized weighted pools. Um which is quite interesting because back in the day we only had 50/50 two tokens that the module popularized by UNISOP v2 but in

balance your v1 you had multiple tokens up to eight custom fees. Um yeah and that's it. Uh the custom weights allowed us for creating liquidity bootstrapping pools which are pools that allows you for fundraising. If you're conducting a token sale, you can change the weights over time and that allows you to raise funds. Uh and last but not least, one other thing built on V1 which is quite interesting because V1 is 5 years old but COMMS we launched it last year. So the technology we built back in the day was quite flexible and allow us to prototype and

create this uh MV capturing AMM which is still live. We then move it for balancer v2 one year later. uh we introduced the singleton fault and balancer was shifting the narrative from being a product to be a platform where people could build new AMMS new products on top of balancer instead of going somewhere and starting from scratch. We introduced it on balancer v2 the boosted pools which allowed it assets to be deposited on landing markets and then earn additional yield. We also had many partners V and V2. So you had gyro ECLPs, you have save finance building FX pools and you have chrome which built

the first implementation of TMS on on balancer and finally balancer v3. Uh we switched to be this platform for programmable liquidity. Uh but we still saw a lot of room for growth. So on balancer v3 we focus on improvement of the developer experience. uh we double down on fungeible liquidity because that's our main uh client or main customer at baler and we focus on you liquidity again bringing back the boosted pools which I just talked uh about speaking of developer experience on balancer v3 you can build cusenms with either hooks so you can addit uh an excs pool type and

add pre or post interactions before swap half your ad liquidity and stuff like that. But you can also build a custom MM from scratch creating another variant and that's pretty easy, pretty flexible because of the developer experience that we have improved it a lot. Going a bit deeper on the on the boosted pools and and how they work actually on balancer v3 and different from from balancer v2 we have 100% boosted pools. So all the assets uh for instance in a stable pool composed of USDC uh and USDT are deposited into a lending protocol.

So you have the symbiotic relationship between the daxes which in this case is balancer and you have on the other side the lending protocols which could be a could be oiler could be morulo or yeah any other like that which supports the ERC 4626 that's the pattern that's the standard we build the the boost and pools on top of that allows us to provide liquidity providers this additional yield they would not get elsewhere and that's actually shared across all pool types at balancer. So actually a weighted pool, a stable pool, a gyroscope pool can be a boosted pool because it's assets can be deposit in a leni protocol and that's all abstracted

away. The complexity is abstracted away by default. Speaking of custom pool types, we launched balancer v3 last year December. Uh and since then we had two partners which already launched uh on balance v3. So we had gyroscope which are super liquid uh post uh made specifically for stable coins and youth bearing assets as repake ETH uh and other liquidity staking uh of ether or any other asset actually. Uh and we have quantum.

Quantum is a quite interesting one because since the beginning since balance v1 we had this idea of generalized weighted pools and index funds uh being on chain and actually quantum is our more recent uh attempt to tackle that market. So quantum they run quantum strategies uh behind the behind the door. They train models and and stuff like that and they create autobalancing pools which are designed to capture token appreciation because if one of your assets is appreciating in value, you don't want to implement loss.

You actually want to capture that value, that asset that's growing in value. And one latest one uh pretty interesting innovation by by balancer is the stable search hook. uh I would claim this is the hook with most liquidity uh on chain and probably the one that has facilitated most of the of the volume if you compare existing existing hooks across defy uh this one is built on top of stable soft uh so the math uh built by curve but in this case we have a dynamic fee uh so it's designed to protect assets from teagging and charging a fee on the trader ers that

are trying to dump uh their tokens into the stable soft pool. Pretty interesting for stable coin issuers and and other projects like that. That's what we have been building so far. We have innovated a lot, but there are still some challenges that we couldn't tackle yet. Uh and and and I would say one of the remaining ones is just increasing the capital efficiency for weighted posts. So we generalize it the XY equals K the NFP2 formula to any tokens to any weights. But in reality that's a basic AMM. You provide liquidity across all the prices. So if

you're familiar with with AMMs you know that you provide liquidity from zero to infinity. That's not actually efficient. Uh so we know certain tokens will not cost or should not cost. you're you're not willing to provide liquidity on a certain range of prices. So someone people from Uniswap then Robinson came up with uh concentrated liquidity uh they launched it uh which was pretty cool and introduced some some effects some some benefits uh for both the liquidity providers and the traders. So for the traders because you now have liquidity concentrated, you have less slippage, you have a better execution

price and for the liquid providers, you now can earn more fees uh per the dollar you invested because you're only you're concentring your position within uh a price range. That's some of the benefits, but of course there are some drawbacks on it. So now if you provide liquidity uh in a certain price range and the price of the asset say if goes out of range then you have to reposition you have to actively manage uh that position that means at the end of the day if the price goes out of range you'll be only holding one of the tokens you have to sell half and then get back into the pool. So that's a

bit cumbersome. uh you as a liquidity provider might be exposed to just time liquidity attacks because on concentrate liquidity amms the swap fees are distributed across the liquidity providers also according to the concentration they're providing liquidity um and there's this kind of attack where if there's a big trade which is moving the price uh there will be a a bot which will add liquidity in a super tight range in order to get all of the fees of that transactions and the liquidity providers who have been providing liquidity for a long time might not get those fees. So that's another drawback. You have the lack of fees compounding. Uh so all the fees, the soft fees accumulated are

consolidated somewhere else and you have to claim and redeposit again. And last but not least, you have nonfungeible positions. So it's hard to incentivize. It's hard hard to compos to to have composability and and stuff like that. some some solutions uh have been around to try to tackle those problems. I guess you might have heard of automated liquidated managers, ALMs. Uh I would say they're still not there. Uh some of them behave as black boxes uh and you don't really know what they do under the hood, what they strategy is. Uh you might have additional costs like the management fees, performance fees. The

rebalance is a pretty interesting point because some of them do all at on rebalance. Uh so they might buy uh the tops and sell the bottoms because of that behavior. And you have other risks. Of course, some of these contracts are upgradable uh not immutable. You don't know about the audits or or stuff like that. What I think or we at balancer think that liquidity providers really want is of course having the capital efficiency. So having concentrated liquidity um but being passive so no need for active management meant no need for rebalancing and shifting the ranges and

finally also funible positions. So they can compose, they can deposit in other protocols and they can incentivize easily because if everyone shares the same position, it's just a matter of distributing incentives proportionally to the amount of tokens they held. That's what we we think that the liquidity providers want. Uh and in order to achieve that, uh I'm pretty excited about introducing reclaim. That's readjusting concentrated liquidity. AMM we have been working at this for the past since the beginning of this of this year January. Um and we want with reclaims first of all have automated automatic adjustment of the

ranges so the liquidity providers can set and forget their positions. They don't have to care more about it. Uh the liquidity the the AMM does the job because it's automated. uh and you have as as we think the liquidity providers want you have liquidity uh fungeible positions there are a few pairs you can configure in order to create such an AMM the first one is just the target price that's the market price at the deployment time uh and then you have two params which are actually mutable so they can be transparently managed by a pool owner/pool manager which can change those pairs

They being uh first the price range. So that's the mean and the max price you're providing liquidity uh within. You have the margin. That's the threshold that triggers range adjustments. And last but very important you have the price shift. So that's how fast the range moves when you're adjusting. That's pretty important. As you might notice, we don't do all add-ons rebalances. So, it rebalances over time in order to avoid those permanent losses that might that you might have on other solutions like other automated liquidity managers.

How it works? Just three images to to keep it simple. So, in the first one, you have a poll in range as you see you have all the all the curves. So the math behind it, it's still XY= K, the same uni U uni V2 formula, but you have a concentration. So you have the the red points there, which is the which is the price range. Right now on the first image, the pool is in range and within the margins. Uh on the second image, the pole is out of range in between the margin and the and the max or the mean uh price range depends on the the

reference. And then once that happens, the pose start shifting what we call virtual balances in order to adjust its price or the price it's willing uh of buying or or selling assets. And then the market will do the work because there there will be an arbitrage opportunity for the for the MV bots or for the uper generally to bring the pool back to the to the market price. Uh and that's what it happening on the on the latest imaging. So the pool is under adjustment. This can happen over and over again and hopefully over the long run if the pool is well designed, if the

pool is well simulated, then you have good parents and the pool will stay most of the time within the margins because that's where you have the most capital efficiency. If you are within the margin and the and the price ranges, then you don't have that much of capital efficiency. You're actually trying to figure out what the price of the market is. overview of of the solutions out there. So you have concentrated liquidity AMM which the token is an NFT so it's nonfgeable on reclaim that's an ERC20 as all LP tokens the BPTs as we know on balancer the efficiency for both I would claim is high again the math is still the same um and the management in one

side you have active in the other side you have passive that does not mean consider liquidity is not good we're building concent liquidity I guess those are different solutions for for different users and reclaims serves best the the passive liquid providers. The future hopefully what we're looking for at balancer is having dow's adoption uh of this model. If you're a DAO or a treasure man manager of a DAO for sure uh the reclaim can be a good option for you. If you're providing if you want to dip liquidity of your own governance token, you can provide it um via reclaims um define integrations

should be super simple because again it's an ERC20 it's funible you can send transfer you can stake hopefully you can even provide this to to other to other protocols to your farming and create composibility on top of it and finally that's just a simplified UX you You don't have to manage ranges. You don't have to worry about it at all. You just, as I said before, just set and forget. Um, and what balancer provides as as a platform um is we provide a pool simulator. So pairs are quite important uh on this case and you have to be very

sure about what params to choose for different markets for different pairs. So we're building internally this post simulator uh hoping to release it and make it open source alongside docs uh probably by the next week. uh we are talking to aggregators because one of the most important thing in order to for this solution to also be profitable for the liquidated providers is to have organic flow so you don't rely all the time on the arbitrageers which are just seeking profit. Uh so we're talking to aggregators and actually we have already got uh two C solvers integrated with it and last you also have uh the pool

creator UI which hopefully when the simulator and the creator are defunded like AC across people then people can use it right now I think those will be more gate cap uh because it's still a nent technology we don't want to make mistakes but it's still it will all be open source and out there for anyone want to try. Last but not least, Balancer cares a lot about security. So, Balancer V3 and Reclams have been audit by the best firms out there. So, we had Spurbit, Tree of Bits, and Sur audits like all the hooks and all the custom pools we create. So, shout out to them.

And we also have this bug bount program with up to to a million dollars uh in Minifi. That's it. Uh thanks for for having me. Thanks for listening. If you want to reach out to me, if you're a DA, if you're a liquidity provider, if you're a treasure manager and want to talk to us, we'll be here. We'll be on ETC as well. We're happy to talk to to you guys. Thank you. [Applause]