Automated market maker exchange Bancor has rolled out a new mechanism that allows users to increase their capital efficiency while providing liquidity in its pools.

Called Vortex, the solution allows users providing liquidity in BNT, Bancor’s utility token, to borrow funds while continuing to obtain yield from swap fees.

The Vortex mechanism reworks the existing mechanism of vBNT, a special version of the BNT token that entitles users to participate in governance. The voting token is automatically received when staking BNT into a liquidity pool, and it can be defined as Bancor’s pool token.

The Vortex proposal adds functionality to vBNT, creating an infrastructure that allows users to sell the token for the original BNT. Once vBNT is converted, users can exchange it into any other asset.

The vBNT sale mechanism makes Vortex a no-liquidation lending platform, letting liquidity providers receive their future rewards immediately, in a similar manner to Alchemix. Since their principal continues to accrue swap fees, the loan will eventually repay itself.

The “no-liquidation” part of the loan comes from the fact that vBNT and BNT are essentially the same token, and the rise in price of the BNT collateral is very likely to be mirrored by vBNT. BNT staking creates vBNT at a one-to-one ratio, but the price relationship between the two is not straightforward.

Combining protocol revenue and lending results in complex tokenomics

The vBNT token’s price is derived from a BNT/vBNT AMM pool, thus largely being defined by the market. A potential arbitrage mechanism means that vBNT is unlikely to ever be worth more than 1 BNT, as arbitrageurs could simply stake BNT, sell the vBNT, and obtain more BNT than they started with. The cycle could be repeated an infinite number of times until the vBNT price returns below 1 BNT.

At the same time, vBNT has no price floor because the arbitrage mechanism cannot work in reverse. As Mark Richardson, the creator of Vortex, explained to Cointelegraph, Bancor uses internal records to define ownership within an AMM pool. This is a significant difference from models like Uniswap’s pool tokens, which are the sole marker of liquidity ownership. The vBNT could be used to redeem a BNT liquidity pool only if that address had already created one.

To guarantee that vBNT maintains some value in the absence of a redemption mechanism, the protocol will be conducting a buyback-and-burn strategy on the token. A governance-defined portion of the protocol’s fee revenue will be diverted to periodically buy and destroy vBNT from the pool with BNT, providing a constant buying pressure.

This has the added result of creating a sink of BNT and vBNT. Since one vBNT unlocks one BNT, destroying vBNT supply creates an imbalance with the tokens contained in AMM pools. A portion of those tokens would thus remain locked in the pools forever, though this should not impact liquidity withdrawal for individual liquidity providers due to the large excess capacity — a similar mechanic occurs with cold wallets on centralized exchanges.

The vBNT token mechanics have a number of interesting ramifications. In addition to the ability to borrow while continuing to receive yield, liquidity providers are also able to leverage their liquidity to receive more swap fees. The price of vBNT directly affects how leveraged the system can be, as prices close to 1 BNT could support an almost infinite leverage factor. At the same time, as more LPs enter leveraged positions, the price of vBNT is likely to decrease and limit the leverage multiplier. An infinite leverage situation would extract value from the protocol, but Richardson is confident that the market-based pricing mechanism quickly makes this costly and ultimately impractical.

Liquidity is no longer an issue, but volume is trailing behind

The Bancor protocol has deployed every resource it has to draw liquidity into the protocol. Between the innovations of single-sided liquidity provision and impermanent loss insurance, introduced with V2.1, it has also launched aggressive liquidity mining programs. The Vortex proposal is yet another tool that could draw liquidity in by introducing leverage on AMM pools.

Bancor’s liquidity campaign has been a demonstrable success. With $1.8 billion in total value locked, it broke into the “billion-dollar TVL club” to become the eighth in the decentralized exchange rankings on DeFi Llama. While it is behind most of its direct competitors such as Uniswap or SushiSwap, Bancor has grown much faster as it started the year at just $140 million in TVL.

The growth in liquidity hasn’t automatically resulted in more volume, however. Though Bancor is in the top-five by volume on Ethereum at $430 million per week, Uniswap dominates the market and attracts almost 17 times as much volume despite only having slightly more than twice the TVL. In Richardson’s view, the Bancor team may have had misguided expectations in its pursuit of liquidity:

“There was this assumption, I would say — and we might not have even been aware that it was an assumption — that if the TVL gets high enough, it will just attract traders [...] And if everyone’s using aggregators, then that’s really good for us because we just have to offer the best product at the lowest rates and traders will just use us.”

The reality turned out to be less idealistic than expected as the team found out. “It turns out no one uses aggregators, and traders hardly ever are using the pools with the best rates,” Richardson added. “They just do whatever they’re going to do.” Nate Hindman, head of growth at Bancor, had his own view of why Uniswap is so dominant:

“I think a big part of that has been this sort of ‘Uniswap gems’ movement that was a DeFi summer thing, where there’s all these new tokens that are launching pools on Uniswap. So, Uniswap is the only place to get these ‘gems.’”

Hindman’s assessment seems to be in line with Uniswap’s volume data. According to its statistics, the volume distribution is heavily skewed toward smaller tokens. Pairs between Ether (ETH), Bitcoin (BTC) and stablecoins take about 25% of the total volume, while the rest of the list is populated largely by low-capitalization tokens that are hard to access on other platforms.

As Hindman revealed, capturing the “long tail of tokens” will be Bancor’s next major objective. One potential proposal for that is the Origin Pool, which allows creating “synthetic” pools paired with ETH, whic is seamlessly replaced with BNT by the protocol. This would solve long-standing onboarding friction for Bancor, as projects wishing to get listed needed to hold BNT in addition to their own token.

After the Uniswap V3 announcement and its heavy focus on swap efficiency — partially at the expense of liquidity pool automation — it became clear that AMM projects are starting to diversify into different niches. With SushiSwap’s focus on additional features such as margin trading, Balancer’s push for composability, and Bancor’s approach focusing on the LP and the BNT token, the AMM space is becoming more and more varied.