The ancestor of AMM, Bancor, makes a comeback. Can it surpass Uniswap by leveraging Layer 2 next?

This article is from Mars Research, authored by Song Qinghua and Mei Ninghang.

Will Bancor be the dark horse of DEX in 2021? If we judge based on the performance of the past few months, we might make such a conclusion; however, if the high gas fees are not resolved, the liquidity gained through effort could also be lost just as easily.

Mars Research is optimistic about Bancor based on our assessment of the trend of DeFi projects like Bancor and Uniswap in laying out Ethereum Layer 2, as this will prompt a reshuffling and restructuring of the entire DEX ecosystem where Bancor and Uniswap reside. As for why we are more optimistic about Bancor's opportunities in this new situation, we will discuss this in detail in the latter part of the article.

Mars Research introduces Bancor to everyone, primarily because of the wealth opportunities that BNT may bring us; but we also want to tell this story in detail—born at the peak, rapidly falling, facing skepticism, staying true to its original intention through numerous iterations, and finally fighting back for a chance to become a BATTLE overlord and reclaim lost ground. We hope this story can inspire fellow travelers in the domestic crypto world.

1. The Situation: Losing Grounds Through Efforts

Headquartered in Switzerland, Bancor raised $153 million from investors like Tim Draper and Blockchain Capital in just three hours, an event that took place in June 2017. That year, the average ICO financing amount in North America was $31.5 million, $30.7 million in Asia, and $16.7 million in Europe.

Why did it receive such great favor from investors? Let's take a look at its ambition: the term Bancor originates from the post-war international settlement currency envisioned by British economist John Maynard Keynes during World War II—the prototype of a super-sovereign currency, which was later replaced by the Bretton Woods system.

Satoshi Nakamoto's BTC originated from the government's excessive issuance during the 2007 financial crisis, leaving economic troubles behind; BTC is a cryptocurrency designed to resist malicious issuance, currency devaluation, and inflation. The Bancor protocol is a way to achieve decentralized exchanges (DEX), and its vision is to solve the liquidity problem of future cryptocurrencies. Therefore, we can see that Bancor's main concept is to provide liquidity:

First, there are cryptocurrency traders, the end users of the Bancor protocol, who can convert tokens through Bancor's pricing algorithm and trade through various channels; second, there are issuers, including individuals, companies, communities, and organizations that can issue cryptocurrencies; third, there are those seeking asset securitization, who can map real-world assets or other on-chain token assets into smart tokens by creating proxy tokens, composite tokens, etc., at a ratio of 1:1. This includes two very important concepts: asset securitization and cross-chain assets; finally, there are statistical arbitrageurs, also known as "brick movers," who are needed by the system and play an important role in stabilizing the prices of tokens within the system. For them, whenever there is a price deviation, the opportunity to make money arises.

After Bancor went live, it was not understood by most people, and there were even some doubts, especially in 2019 when it was forced to prohibit U.S. users due to uncertainties in SEC regulation. We can only say that circumstances create heroes—although it succeeded in the ICO, in the larger trend, DEX was still a future concept, and during this period, DEX had not yet taken center stage.

Later, Uniswap's "automated market maker" (AMM) and constant product market maker (CPMM) were both based on Bancor's pioneering practices. Automated market makers (AMM) use algorithms called "Money Robots" to simulate price behavior in markets like DeFi; constant product market makers (CPMM) were promoted in the first AMM-based DEXs, Bancor and Uniswap.

CPMM is based on the function x*y=k, which determines the price range of two tokens based on the available quantity (liquidity) of each token. When the supply of X increases, the supply of Y must decrease, and vice versa, to keep the product k constant. When the curve is plotted, the result is a hyperbola, where liquidity is always available, but as the price increases, the ends approach infinity.

We must admit that Bancor is the ancestor of AMM in the crypto field; we also have to acknowledge that this model became popular due to Uniswap, which expanded it. An objective analysis shows that part of the reason is that Uniswap's simplicity indeed helps more people participate, providing a better user experience. But more importantly, Bancor was banned just as Uniswap emerged, riding the wave of DeFi projects like Synthetix in the 2020s; the AMM ancestor Bancor led an era but thus missed another era.

2. Iteration: Opportunities Regained Through Efforts

After Uniswap made some improvements and strongly occupied the DEX market, the Bancor development team successively launched Bancor V2 and Bancor V2.1, addressing the concerns of market makers regarding impermanent loss, liquidity token exposure, and traders' concerns about whether slippage is low enough.

01 Taking oracle pricing AMM as an example, Bancor V2 broke the traditional AMM model, allowing the values of the two tokens not to be equal. V2 uses oracle pricing to adjust the weights of the two tokens, meaning that the quantity of token A * A price does not have to equal the quantity of token B * B price, smoothing out arbitrage opportunities with oracles.

02 Using stable curve as an example, Bancor V2 drew inspiration from Curve, with Bancor's curve lying between Curve and Uniswap, algebraically between X*Y=K and X+Y=K. By utilizing a smoother curve, it reduces slippage. Simply put, it magnifies the liquidity under the same conditions by 20 times compared to Uniswap.

03 Based on these two, let's analyze how Bancor V2 avoids impermanent loss. Bancor V2 uses oracle pricing to drive away arbitrageurs, with the key point being that it allows AMM to be unbalanced and adjusts asset ratios in real-time.

After BNT's price increased fourfold, Uniswap had to be arbitraged by arbitrageurs for ¥10,000, bringing the asset ratio back to 50%/50%, while Bancor, which introduced oracle pricing, perfectly avoided being arbitraged and simply adjusted the asset ratio according to the price.

04 Bancor V2.1 also extended its competitive reach to lending protocols. To address impermanent loss, Bancor V2 can provide liquidity for a single token. Through Bancor V2.1, liquidity providers can now pledge single-sided assets and earn returns of 60-100% with complete impermanent loss protection.

The left image is from Uniswap, where users providing liquidity for the LINK-WETH pool need to inject 50% LINK and 50% WETH. The right image is from Bancor, where users can deposit their LINK into the BNT-LINK liquidity pool to provide liquidity for that pool, thus earning liquidity fees from the pool and newly issued BNT rewards (this mentions the incentive role of BNT in the ecosystem, which will be elaborated on in the next section regarding the Bancor team's design that we previously underestimated or even questioned). This is essentially similar to crypto lending services, similar to users depositing LINK into Aave or Compound.

Since the launch of Bancor V2.1, the locked amount has increased by 70 times, and trading volume has increased by 55 times. Some have judged that Bancor is the most promising dark horse of 2021. However, this judgment clearly overlooks the most critical issue troubling all DEX projects—high gas fees.

In the past period, the main motivation for attracting liquidity providers to participate in Bancor has been its exclusive single-token mining and impermanent loss protection plan. However, these users will soon drift away due to Ethereum's poor efficiency and high contract fees. For example, the following image shows this even when the network is not congested.

In fact, high gas fees have troubled all DEX project developers, including Uniswap, thus limiting user participation. According to Dune Analytics data, the number of Ethereum DeFi users has reached millions, but compared to the total number of cryptocurrency users, this is a small portion. According to a new report released by Crypto.com, the total number of global cryptocurrency users increased from 66 million in May 2020 to 106 million in January 2021.

The development of DeFi has already surpassed the carrying capacity of the Ethereum Layer 1 ecosystem, and Vitalik Buterin has become like an ant on a hot pan. Therefore, he has been busy promoting the development of Layer 2 and urging miners to lower gas fees. In this broader context, Bancor and Uniswap have chosen relatively mature Arbitrum Rollup and Optimistic Rollup solutions, respectively.

If their plans go smoothly, in the coming months, Bancor and Uniswap will be able to achieve migration to Ethereum Layer 2, which will mean that the biggest constraint on the entire DEX industry will be resolved, and the number of DeFi users will leap from millions to tens of millions—we predict that a new battlefield will open. Yes, this is a new battlefield because, prior to this, high gas fees led to DEX users being primarily large holders, so even the leading Uniswap and Sushiswap have not established strong moats, and Bancor does not need to worry about user inertia at all.

It can be said that in this new phase, they are at the same starting point. This is crucial for Bancor, as it differs from many projects that have developed from the ICO era but have long since died out, run away, or are operating poorly. This is an opportunity that the Bancor team has continuously iterated and fought for despite facing numerous doubts and challenges.

So, since they are at the same starting point, why do we judge that this will be an opportunity for Bancor?

3. Value: Understanding BNT at Last

In V2.1, in addition to the aforementioned provision of liquidity for a single token, there are some changes based on V1 worth analyzing: LPs only need to put one token into the liquidity pool, while the other half of the liquidity pool is provided by the Bancor protocol and exists in the form of newly issued BNT (Bancor protocol token); fees are not entirely given to LPs, as LPs only invested half the value in the liquidity pool, so they only receive half of the transaction fees, while the other half goes to the Bancor protocol, and these fees will ultimately be burned upon withdrawal; based on this, the most profound change is that Bancor compensates LPs for any impermanent loss by minting new BNT to cover the losses.

Undoubtedly, these settings will activate a large number of users from the crypto world to join Bancor, and the performance in the four months since the launch of V2.1 has proven this. Until now, the advantages of Bancor's token model have only begun to be recognized—by integrating BNT as a base asset, LPs and token holders have the same vested interests, as BNT token holders will also become single-sided liquidity providers, and impermanent loss will thus become a thing of the past. BNT token holders can earn yields of up to 50-100% without impermanent loss, providing a huge incentive for LPs to stake their BNT tokens. Since the V2.1 version, the amount of staked BNT on the Bancor platform has surged, currently accounting for 61% of the total staked tokens.

It's time to review and understand Bancor's token economic model. The explanations and clarifications regarding the Bancor protocol's calculations are very clear in Bancor's white paper and formula calculations, with the core being to explain the price determination mechanism of CW.

Many articles and interpretations focus on analyzing Bancor protocol's formulas, calculations, and the CW fitting curve, as this is the core of the price mechanism and is crucial for making profits from trading tokens. However, if one wants to understand the Bancor protocol from a higher dimension, merely understanding the price calculation formula remains at the level of "technique."

To understand the core of the Bancor protocol, one must first grasp the important position of "liquidity" in finance. All use cases and algorithms of Bancor ultimately revolve around various dimensions of liquidity depth, cost, and implementation difficulty.

Thus, in the earliest Bancor protocol, BNT played the role of an intermediary currency in this ecosystem—to automate its AMM services, Bancor aims to incentivize users to deposit assets into pools, with each pool consisting of a pair of tokens and a reserve of BNT cryptocurrency. When users deposit coins into the pool, they receive a new token as a reward. This token is called a pool token, which allows users to retrieve their original amount locked in the protocol. When each token is traded, the BNT token is used as an intermediary currency. Notably, Bancor allows users to lock one token in its pool (rather than a pair). For example, in a pool consisting of ETH and DAI, users can only deposit ETH or DAI. In Uniswap, as an option, users must deposit both ETH and DAI simultaneously. However, users must also deposit BNT into any Bancor pool.

This setting of BNT in the Bancor protocol was referred to as the "token exchanger" model by early researchers in the crypto world, issuing BNT backed by ETH and then using BNT as a connector to issue other tokens. They believed that this does not fundamentally solve the liquidity problem of marginal varieties, and the value of BNT lacks effective support, indicating that this model may have room for improvement and development in project business expansion.

Even supporters could only vaguely evaluate that the governance token BNT participates in the operation of the protocol's economic bottom layer, with a stronger value capture function. However, after multiple iterations by the Bancor team, Bancor V2.1 has begun to unleash the vitality of BNT in this ecosystem, and the Bancor Vortex mechanism makes the role of BNT in the ecosystem stand in stark contrast to that of Uniswap's governance token UNI.

The Bancor Vortex mechanism allows BNT assets to be staked in whitelisted asset pools, borrowing pool tokens vBNT, and supports using vBNT to swap for other tokens in bancor.network. By selling vBNT, users can enhance capital utilization, such as leveraging liquidity provision to earn more fees and BNT rewards.

The Bancor Vortex introduces governance in two parts. BIP9 generally discusses leverage functions and proposes a vBNT burning mechanism; importantly, its unique whitelisted status has been approved, allowing vBNT liquidity providers to earn transaction fees, thus benefiting from the Vortex without utilizing their staked BNT.

In this model setup, based on data up to March 2021, BNT stakers can earn the highest exchange fee income among all DEXs.

However, there are risks associated with using Vortex to borrow staked BNT. Simplifying all risks into a simple idea: what if the cost of going back and forth exceeds the opportunity to make money? More accurately, the total selling price of vBNT + leveraged profits must at least equal the repurchase cost to achieve a break-even point.

Therefore, returning to the fundamental position stated at the beginning of this article, for Bancor, this is not enough to win. The key to winning with this model setup for Bancor lies in a significant improvement in Ethereum network efficiency; otherwise, high gas fees will cause these LPs to quickly drift away. Once Ethereum's network efficiency improves and gas fees decrease significantly, the value of BNT in the Bancor protocol will be released, and the DEX landscape will be reshaped.

The good news is that Vitalik and Rollup developers are steadily promoting Ethereum Layer 2 and lowering gas fees; Bancor and Uniswap are both accelerating their respective Layer 2 Rollup solutions. To summarize, Mars Research's optimistic judgment on Bancor is not based on the performance since V2.1 and the Bancor Vortex, but on the advantages of this model in the Layer 2 era.

We predict that the ancestor of AMM, Bancor, has the opportunity to surpass Uniswap through Layer 2. So, we seem to need to face an open topic together—Is Ethereum Layer 2, which has Vitalik anxious, reliable? We will specifically discuss the pros and cons of the four mainstream Rollup solutions in future articles.