Delphi Labs: New Token Issuance Mechanism, Locking + Liquidity Guided Auction

Written by: Delphi Labs

Compiled by: MIM, Deep Tide

Introduction

Delphi has invested a significant amount of time in analyzing and designing token issuance mechanisms. Overall, there has been substantial progress in token issuance mechanisms since the fixed price sales of ICOs, but we still believe there is considerable room for improvement.

We recently incubated two projects, Mars and Astroport, but while planning their token issuance mechanisms, we found that none of the existing structures could fully meet all our design goals. To achieve a "tailor-made" solution for our design objectives, we reworked the token issuance mechanism from first principles. The result of this modification is the "Lockdrop + Liquidity Bootstrap Auction," which will be trialed in December on Astroport and subsequently on Mars.

This article will explain in detail the design goals we are trying to optimize and why we believe the new structure is the best token issuance mechanism to achieve these goals.

Design Considerations

Current Token Issuance Mechanisms

Broadly speaking, the goal of token issuance mechanisms is to distribute tokens to the users and community of the protocol. The various current projects primarily adopt the following two token issuance mechanisms:

(1) Distributing tokens to users—users receive tokens based on past actions or ongoing actions. This mechanism includes airdrops and all forms of continuous incentives (staking rewards, liquidity mining rewards, trading competitions, etc.).

(2) Publicly selling tokens—users receive tokens in exchange for investment money. This mechanism includes fixed price sales, auctions, LBP, Pylon-style yield delegation, etc.

Both mechanisms have multiple flaws.

Distributing Tokens to Users:

The protocols that distribute tokens to users mainly adopt two methods, namely airdrops and continuous liquidity incentives.

Airdrops aim to allocate tokens based on users' past behaviors, either because they have previously used the protocol or because they are high-value users of other protocols. This seems like a good idea, but airdrops reward past behavior rather than future commitments, so they cannot guarantee benefits for the protocol's future development. In fact, our research shows that most airdrops are later abandoned.

Continuous liquidity incentives aim to reward ongoing participation in the protocol and have become the preferred distribution method for most projects. However, protocols relying solely on this method also have some drawbacks. First, continuous incentives can only reward usage that can be reflected through on-chain behavior, such as providing liquidity, collateral, or completing transactions. This benefits whales with more capital to gain a higher share of rewards, but it also excludes other stakeholders who add value to the protocol, such as community members and third-party integrators, and even projects built on top of the protocol. For these stakeholders, acquiring tokens on the open market may be the only way to achieve incentive alignment.

Second, simply distributing tokens to suppliers can lead to additional problems, such as low initial float, lack of price discovery mechanisms, and low liquidity (especially in the early stages). This can harm non-suppliers—those who intend to purchase tokens on the open market may either be unable to buy in scale or face price fluctuations due to low liquidity (such as the launch of ANC) or the continuous selling pressure from low-float token emissions (such as MIR), resulting in a double whammy.

Publicly Selling Tokens

To prevent the aforementioned issues, some projects choose public sales to facilitate price discovery, but public sales themselves also have drawbacks. The first and most important point is that all types of public sale mechanisms expose projects to higher regulatory risks. Involving obvious "investment of funds" and early-stage protocols is most likely to be deemed a "basic entrepreneurial effort" dependent on a small group under the "Howey Test." This factor, along with others, leads to public token sales being more likely to be viewed as unregistered securities sales compared to other distribution methods.

Second, while public sales address the price discovery issue, they do not necessarily solve the problem of low initial liquidity. Although projects can increase initial liquidity themselves, this requires a significant amount of capital and exposes them to regulatory risks, as public sales essentially set the price for tokens in the market. Therefore, many tokens exhibit high volatility in the days and weeks following their launch—crucially, this high volatility often harms early users who are enthusiastic about the project, as they may not be sufficiently skilled in trading.

Third, most sales and auction structures are susceptible to front-running attacks by bots (e.g., ANC, VKR), leading to supply concentration in the hands of a few whales with technical capabilities and capital, allowing them to dominate the entire sale.

Summary: Design Goals and Constraints

In summary, we aim to propose a token distribution mechanism that can achieve the following goals:

Distribution: We want to ensure that tokens are fairly distributed to a wide range of stakeholders. Most importantly, this means that the launch mechanism should not be exploited by bots or whales. Ideally, we want to give smaller holders an opportunity to earn more tokens by demonstrating their commitment to the protocol.

Price Discovery: We want to ensure there is a fair, decentralized, bottom-up mechanism to establish the token price before token trading.

Sufficient Float: Relatively, we want to ensure there is enough initial supply to meet demand and provide assurance for legitimate price discovery. Recently, many projects have launched with extremely low float (less than 5% of total supply), leading to highly inflated fully diluted valuations (as demand is somewhat inelastic, there simply isn't enough supply to meet this demand). From here, continuous token emissions and unlocks lead to a chronic "bleeding" of its price over time, causing community members to become dissatisfied and lose motivation. We firmly believe that the only way to create a successful protocol is to involve the community from the very beginning, allowing them to grow and succeed alongside the project. For this reason, a key consideration is to ensure that the initial float is sufficiently large to achieve fair and legitimate price discovery.

Liquidity: Once the price is determined, we also want to ensure that the determined price has immediate and deep liquidity.

Decentralization: The joint venture participating in the creation of Mars and Astroport does not wish to set prices, conduct public sales, initiate AMM/LBP, accept "venture capital" investments, or otherwise position itself as a seller, authority, market maker, or broker.

Therefore, it is crucial that users of Astroport provide all liquidity and price discovery on a decentralized basis.

Lockdrop + Liquidity Bootstrap Auction

We believe that "Lockdrop + Liquidity Bootstrap Auction" can achieve all these goals. At the highest level, it is a two-phase process that works as follows:

Phase One (Lockdrop): Allocation Phase. This phase provides a time window during which anyone can pre-commit to becoming a user of the protocol within a certain timeframe (specific details are provided in the "Phase One: Lockdrop" section below).

At the end of the time window, participants will receive a proportional share of the total tokens being distributed based on the size and duration of their commitments. These tokens will remain locked until the end of Phase Two.

Phase Two (Liquidity Bootstrap Auction): Price Discovery Phase. This phase provides a time window during which participants from the lockdrop phase can commit to providing liquidity by depositing part or all of their ASTRO into one side of a stablecoin liquidity pool (i.e., ASTRO-UST). Then, other users can join in and deposit their UST into this liquidity pool, effectively purchasing ASTRO from the lockdrop participants. At the end of this phase, the following will occur:

1. Native tokens + stablecoins are deposited into the liquidity pool. The ratio of native tokens to stablecoins determines the final price of the native tokens.

2. Auction participants receive LP shares proportional to their deposits, which are locked and vest linearly over three months. Assuming sufficient participation in Phase Two, this will ensure that the market price has immediate and deep liquidity.

3. All tokens distributed in Phase One that are not used for liquidity auction commitments in Phase Two will be unlocked and available for free trading.

Note: Tokens can also be distributed to users via airdrops in Phase One (either as an alternative or as a supplement to the lockdrop).

To illustrate how this mechanism works in practice, we will examine this issue in the context of "Astroport's Token Issuance Activity":

Phase One: Lockdrop

Phase One is the distribution phase, aimed at delivering tokens to Astroport users. Ideally, the number of tokens users receive should be proportional to the "value" they bring to the protocol. The measurement of value varies, and different distribution mechanisms correspond to different forms of value. For example, the airdrop mechanism recognizes the value brought by historical users of the target protocol or historical users of synergistic protocols likely to use the target protocol. The lockdrop can be seen as similar to an airdrop, but it rewards users for their forward-looking commitment to using the protocol in the future rather than rewarding past actions—it is akin to receiving discounts when signing a one-year contract with your mobile or broadband service provider.

For Astroport, these long-term protocol customers are liquidity providers who can guide liquidity in key trading pairs. During the lockdrop period, users can commit to depositing liquidity in the form of Terraswap LP shares into Astroport for pre-selected trading pairs. Crucially, users can further demonstrate their commitment to Astroport by setting a long lock duration for their liquidity; the longer the lock period, the higher the rewards, with a maximum lock period of one year. 7.5% of the total ASTRO supply will be allocated to users who lock liquidity in Phase One, hence the name "lockdrop."

Overall, the "lock" share received by an individual user will be calculated based on two factors:

1. The total ASTRO allocated to the trading pairs contributing liquidity to the user (to be announced before the start of Phase One).

2. The user's weighted adjusted liquidity share in a pool. It is not merely a measure of the user's liquidity or total liquidity but includes a weight adjuster based on the time the user locks their LP tokens. This means the numerator is the user's weighted adjusted liquidity, and the denominator is the total weighted adjusted liquidity provided by all users in that pair.

The ASTRO share a user receives for their LP contribution to a trading pair is calculated as follows:

A = Ap * (wLu/wLp)

A = ASTRO received by the user

Ap = Total ASTRO allocated to the user's chosen trading pair

wLu = User's weighted adjusted liquidity

wLp = Total weighted adjusted liquidity in the pool

B = Boost value based on the selected lock duration

According to the schedule, users can freely deposit and withdraw during the initial 5 days to encourage anyone to participate and invest all the liquidity they are willing to provide. Users can only withdraw during the last 2 days if they feel the amount of ASTRO they receive from the lockdrop is less than the opportunity cost of that liquidity; the system allows them to withdraw part of their liquidity.

At the end of Phase One, liquidity will be migrated from Terraswap to Astroport—users' Terraswap LP shares will be burned in exchange for locked Astroport LP shares.

Importantly, in addition to the one-time lock rewards, LPs will also receive trading fees + issuance fees from their pool (as will other liquidity providers).

In addition to the 7.5% ASTRO allocated during the lockdrop, another 2.5% ASTRO will be distributed via airdrops to LUNA subscribers and Terraswap users. This provides 10% of float available for Phase Two.

Phase Two: Liquidity Bootstrap Auction

Phase Two is the price discovery phase, aimed at establishing a price for the tokens and ensuring deep liquidity based on that price. For those familiar with trading, the liquidity bootstrap auction can be seen as similar to recent auctions on Mango Markets, but there are several key differences:

1. Users are selling tokens, not the protocol;

2. Auction participants receive locked LP shares, not unlocked tokens;

3. In the last two days, withdrawals will be gradually restricted to prevent whale manipulation (this will be explained in more detail below).

In Phase Two, users who participated in the lockdrop or received tokens via airdrop can choose to deposit part or all of their locked ASTRO tokens into the ASTRO-UST liquidity pool. Then, any user can join in and deposit UST on the other side, effectively purchasing ASTRO from the locked participants.

The price of ASTRO is determined by the amount of UST and ASTRO added to the pool, specifically the final ratio of UST to ASTRO. If 100 ASTRO and 100 UST are deposited, the implied price of ASTRO is $1; if another 100 UST is deposited, the implied price jumps to $2. The key is that this price is meaningful because participants commit to initializing and locking their tokens in the liquidity pool for three months at this price (i.e., they are effectively buying ASTRO for less than this price and selling it for more than this price within three months)—this is a typical risk faced when providing liquidity in an AMM, where they cannot spontaneously exit based on market conditions.

According to the schedule, during the first 5 days of Phase Two, users can deposit as much ASTRO and UST as they want while only being able to withdraw UST. This encourages users to participate and bid the maximum amount they are willing to invest. Users can only withdraw UST in the last 2 days. As UST is removed, the implied price of ASTRO will decrease. This makes price discovery possible, as users will continue to withdraw UST until they see the ASTRO:UST ratio reflecting an acceptable market price for ASTRO.

The events of the last 2 days are worth examining in detail, as other similar designs face the risk of manipulation during this phase. Once it reaches Day 6, the withdrawal limit will be capped at 50% of the deposit amount. By Day 7, the withdrawal limit will linearly decrease from 50% to 0 over the course of one day. This limitation is necessary; without it, whales would over-invest (i.e., invest significantly more than they plan to leave in the pool), trying to push the price far beyond what they actually want to pay, thereby preventing others from participating. Then, when the time is up, they would withdraw all the excess funds, locking in a lower price. Such manipulation has occurred on Mango Markets before, severely undermining token issuance and price discovery. The withdrawal cap can limit the effectiveness of manipulation, and the gradual decrease creates increasing stability around price discovery, leading to closure.

At this point, the mechanisms of the last few days of Phase Two should be quite clear, and we need to continue exploring the reasons for broad user participation in Phase Two. Whether they prefer to sell their rewards or use them for LP, they can do both without participating in Phase Two while retaining the option. It is called retaining the option because users' LP shares will be locked for 3 months when they participate in Phase Two. In response, we will provide rewards for participants in Phase Two, in the form of an additional 1% ASTRO supply (10 million ASTRO). This 1% will be distributed between ASTRO holders and UST holders, effectively serving as a premium for sellers and a discount for buyers. The fixed 1% reward means that if participation in Phase Two is low, the bonus will increase relatively. This helps create secondary market liquidity, ensuring it reaches the participation level we deem necessary. We present below a scheme for distributing rewards to Phase Two depositors based on overall participation rates:

While we list all participation scenarios in the table, we certainly do not expect participation rates to approach 100%. Below, we provide an example with integers to illustrate the application of the table.

For example: 25% of the Phase One token holders decide to participate in Phase Two (25,000,000 ASTRO tokens), and 25,000,000 UST are also deposited. This would result in a price of $1 per ASTRO. These participants would receive an additional 10,000,000 ASTRO (5 million each) based on their deposit proportions.

Assuming you hold 1% of the deposited ASTRO, you would be entitled to receive 1% of the 5% ASTRO reward. This means your deposit of 250,000 ASTRO now represents LP tokens of 125,000 ASTRO and 125,000 UST, allowing you to receive an additional 50,000 ASTRO, which is up to a 20% reward.

At the end of Phase Two, auction participants will receive LP shares proportional to their contribution of liquidity, which will be locked and vest linearly over three months.

At this point, the circulating supply of ASTRO will reach 11%.

Advantages

We have detailed the new mechanism, and next, we will explore how it achieves the goals outlined in the "Design Considerations" section.

Distribution: ASTRO is allocated to users who pre-commit to using the protocol. Small investors can increase their ASTRO share by voluntarily locking in a longer usage period, while non-users can obtain tokens in a way that is not exploited by whales or bots, allowing users some control over the price they pay.

Price Discovery: Phase Two implements a bottom-up price discovery auction mechanism through LP shares, allowing buyers (users committing UST to LP) and sellers (Astroport users committing ASTRO to LP) to jointly determine the fair price of ASTRO.

Sufficient Float: By the end of Phase Two, there will be 110,000,000 ASTRO in circulation, equivalent to 11% of the total supply.

Liquidity: Phase Two ensures liquidity through the auction of locked ASTRO-UST LP shares rather than auctioning the ASTRO tokens themselves. The liquidity pool will be automatically funded at the end of Phase Two to ensure it has immediate and deep liquidity.

Decentralization: The project team or DAO has not sold tokens to the public. Instead, protocol users are granted tokens (essentially for user-based governance rather than as part of an "investment"), and they can then choose to sell part of them through the auction mechanism. Additionally, due to the significant float, the potential impact of pre-issuance builders on post-issuance token-based governance is diluted, allowing for decentralization of governance to occur much earlier than usual.

Conclusion

We believe that "Lockdrop" and "Liquidity Bootstrap Auction" represent new fundamental elements of token issuance, and we are pleased to see the broader cryptocurrency community choose to adapt and build them. While we have chosen to combine the two, they are both independent mechanisms that can also be used separately.

"Lockdrop" can be seen as a new method for distributing tokens to users, rewarding future commitments rather than past or ongoing actions. It can be used for any protocol involving "usage" and capital commitments that can be tokenized and locked. The "Liquidity Bootstrap Auction" proposed in this article does not strictly fall under the categories of token distribution or public sales; it can be said to represent a new type of token distribution: peer-to-peer auction. In other words, for those relatively able to tolerate regulatory risks, the "Liquidity Bootstrap Auction" can purely serve as an excellent price discovery and liquidity mechanism, one that involves tokens sourced from the project’s treasury rather than from users.

We have made the code for "Lockdrop" and "Liquidity Bootstrap Auction" publicly available. We have discussed with several projects interested in utilizing these mechanisms and will work closely with them to help them design according to their goals. If you are interested in applying these mechanisms to your own project, please feel free to contact our team, and we would be happy to help you think through how to realize your vision.