Now that the first ever token offering via Balancer Liquidity Bootstrapping Pool (LBP) is complete, we’re excited to share details on how it went and how it compared to other recent Initial DEX Offerings (IDOs).
When the Perpetual Protocol team launched their PERP token LBP last week, it achieved qualities that have been highly sought-after but rarely found in IDOs — fair competition between humans and bots, minimized volatility, even token distribution among participants, and a healthy and stable price discovery process.
These feats were made possible by the unique features of LBPs, a type of Balancer smart pool, which I’ll break down in this article. If you’re unfamiliar with LBPs, start by checking out Balancer Labs CTO Mike McDonald’s piece.
Key stats from the PERP LBP
Total volume: $21.86MM
Duration: 20,850 blocks; ~3 days
Opening price: 1.60 USDC
Lowest price: 1.058 USDC
Highest price: 2.304 USDC
Average price: 1.635
Closing price: 2.076 USDC
Number of participants: 1,355 (1,221 final token holders)
Key charts from the PERP LBP
Why Perpetual chose Balancer
In looking for the right platform to launch the PERP token, the Perpetual Protocol team identified 3 core principles that the offering needed to abide by:
- It had to give humans a fair chance to participate, despite DEX token sales typically being dominated by bots.
- It had to enable better price discovery, allowing the market an opportunity to naturally determine the fair value of the token.
- It had to be permissionless and scalable.
In their search process, the team weighed several available solutions but found Balancer’s LBP to be the ideal structure, for 3 main reasons:
1. Bots have no advantage over humans.
A typical IDO creates an (unintentional) incentive for participants to be the first to get their orders through, spurring a vicious gas war. This happens because teams usually start the IDO at the price of the previous round. Most of the time that price is clearly below what the market would price the token at the IDO launch, since the project evolved a lot since then. But Balancer LBPs flip this into a game in which participants compete to be the first to identify when the token has reached its fair value, and to place their order when this point is reached.
2. LBPs last as long as the team wants, allowing sufficient time for true price discovery.
This mechanism serves to prevent the FOMO-driven erratic behavior that we’ve become accustomed to in token sales (unfortunately). As seen above in the price chart for the PERP LBP (first graphic at the top), the market managed to find price equilibrium around $2.00 without experiencing wild fluctuations.
3. Participants are free to use any well-known token to purchase the token.
This makes the token sale more accessible to the community regardless of which tokens they hold in their wallets, rather than being limited to just one or a handful of options. This is ensured by two Balancer features: Smart Order Routing (SOR) (recently enhanced with multi-path routes) and the ability for Balancer pools to support up to 8 tokens. So the Perpetual team could have sold PERP directly for USDC and ETH in the same LBP. Even when choosing to have only USDC, Balancer’s SOR ensures that users can purchase PERP using many different tokens that have liquidity with USDC in other pools.
How Balancer LBPs stack up against other solutions
- Uniswap v2
Initial Uniswap Offerings (IUOs) offer teams the benefit of accessing the most liquid DEX market, as well as the most simple and straightforward process for participating in a token sale.
However, these benefits come with downsides.
Since Uniswap pools are strictly weighted 50/50 between the token being offered and ETH, teams using an IUO to launch their token must provide a sometimes prohibitively large amount of capital upfront, equivalent to the total initial value of the tokens they are selling — but not all startup teams have this kind of capital laying around.
But I would be remiss if I didn’t mention the frontrunning and FOMO buying that occur on IUOs.
Since IUOs operate on a bonding curve, once the token sale starts, each purchase of tokens necessitates that the price increases. This creates incentives for investors to be the first to buy, and to acquire large amounts of tokens, at all costs. How is this done successfully? With bots that submit lightning fast orders as soon as the token sale begins, at times paying astronomically high gas to ensure their transactions are mined ahead of anyone else’s.
The result? The token price often immediately explodes, gas prices skyrocket, the upside from token appreciation gets highly concentrated among the most advanced and well-capitalized players, and the token sale opportunity can end before many earnest investors get to participate.
If you’re not tech savvy or rich enough… good luck!
But Balancer pools are… different.
LBPs are fully customizable based on a team’s needs, allowing them to pair their token with any set of ERC-20 tokens of their choice, and to weight the value of the pool among the tokens as they see fit. Those weights can be adjusted periodically based on a predetermined schedule configured by the team, continuously changing the pool’s total value distribution among the tokens.
A) The LBP can launch with a heavily weighted distribution towards the team’s token, requiring a significantly lower amount of capital upfront. This lowers the barrier to entry significantly for early stage startups.
B) Not only can the team select the tokens against which they would like to sell the project token, but Balancer’s SOR also allows investors to pay with any ERC-20 token that has sufficient liquidity on Balancer (there are currently 117+ tokens listed on Balancer).
C) The LBP’s continuous weight adjustment creates downward pressure on the token price, preventing it from skyrocketing out of control and giving investors a chance to only buy at a price they individually consider fair. This creates an incentive for investors to WAIT before they buy, rather than FOMOing in, as the best price will most likely come later during the token sale, NOT in the very beginning.
The Perpetual team launched their PERP/USDC LBP with an initial weight of 90/10, meaning that only 10% of the total value of the pool had to be provided by the Perpetual team in USDC — a real game changer. Over the course of 3 days, the token weights were incrementally adjusted until they reached a ratio of 30/70.
A sophisticated user (apparently leveraging some automated tools, judging by the timeliness of their transactions) actually got rekt by the LBP when attempting to play the type of game that works on IUOs. Expecting price to jump quickly, the user purchased PERP tokens with a very high gas fee immediately once the LBP was deployed, only to sell them back at a loss 20 minutes later. As intended, the LBP’s design caused the token price to move down from its starting price. Count that as a win for humans!
Mesa is a DApp built on the Gnosis Protocol which has been utilized to launch several token sales, including those for MTA, DMG, and DIA.
Using Gnosis’ ring trading feature, Mesa allows investors to participate in a token offering using any ERC-20 token listed on Gnosis.
Mesa takes an innovative approach to token sales, allowing buyers and sellers to place sealed buy and sell orders in advance before the sale begins, and asynchronously matching orders via batches of 5 minute mini-auctions.
This approach is designed to mitigate gas wars from occurring at the time the sale starts, allowing participants to compete on price alone.
Despite Mesa’s intentions and efforts, token sales on their platform are subject to some of the same drawbacks as on Uniswap. In practice, the 5 minute batching of auctions creates a powerful incentive for bidders to get their orders included in the first 5-minute batch of a token sale, which means that well-capitalized automated traders can still gain an upper hand.
In other words, humans may still struggle to out-compete bots on Mesa and gas wars are still not necessarily averted by Mesa’s novel design. Additionally, bidders in a Mesa token sale are never sure whether their bids will be included when orders are later settled.
DIA’s recent token offering on the Mesa platform in August 2020 ran into multiple issues, but I’ll mention one of them. While investors were able to purchase the DIA token using either USDC, DAI, or ETH, they experienced severe spread differentials up to 24% between the DIA/USDC pair and the DIA/ETH pair. This occurred due to insufficient ETH liquidity on the Gnosis protocol, leading frustrated users to threaten legal action.
While Mesa is limited by the liquidity of Gnosis, which has $8MM total value locked (TVL) as of the writing of this post, Balancer LBPs are supported by the liquidity of Balancer, a top 3 ranked DEX with around $750MM TVL (over 93x more). This ensures that investors in LBPs can spend many different tokens to participate, with reduced slippage.
LPBs also provide investors with instant feedback, so when a user submits an order, they can be sure it will be executed at that price, provided a suitable gas price was chosen.
3. Centralized Exchanges
An Initial Exchange Offering (IEO) leverages the high liquidity, visibility, familiarity and volume of a centralized exchange such as Binance. However this would require sacrificing core principles of truly decentralized networks, such as being fully permissionless. On top of that, CEXs may also require significant capital to cover exorbitant listing fees.
We anticipate LBPs will become an increasingly common solution
Given the crucial role of token distribution in the long-term success of any crypto project and the deficiencies of today’s most popular solutions for token offerings, interest in LBPs is likely to grow. PERP’s successful launch is evidence that the promise of LBPs is real.
And there’s more…
LBPs are only one of several use cases for Balancer smart pools. We’ll outline more in a forthcoming post.