VC Perspective: How to Solve the Dilemma of "High FDV, Low Circulation"?

CN
链捕手
Follow
1 year ago

Written by: Ro Patel, Partner at Hack VC

Translation: Azuma, Odaily Star Daily

Current Status of Token Lock-up Design

In the current market cycle, the issuance of tokens with "high FDV, low circulation" has gradually become a mainstream trend, which has raised concerns among investors about the sustainable investment potential of the market. It is expected that by 2030, a large number of tokens will gradually unlock in the cryptocurrency market. Unless demand experiences significant growth, the market will have to bear the potential selling pressure.

As per historical practice, contributors to networks/protocols (including teams and early investors) usually receive a certain proportion of tokens as a reward, and these tokens are locked up according to specific time-based structures. As the primary driving force behind the early development of networks/protocols, contributors should indeed receive appropriate compensation, but attention should also be paid to balancing the interests of other stakeholders, especially the interests of token investors in the open market after TGE.

The proportion design here is crucial. If the proportion of locked tokens is too large and affects the token's available liquidity, it will have a detrimental impact on the token's price, thereby harming the interests of all holders. Conversely, if contributors do not receive appropriate compensation, they may lose the motivation to continue building, ultimately also harming the interests of all holders.

Classic parameters for token lock-up include: allocation ratio, lock-up time, unlocking duration, and delivery frequency, all of which only operate in the time dimension. Considering the current situation, using only the classic parameters limits our imagination space for solutions, so it is necessary to introduce some new parameters to explore new possibilities.

In the following text, I suggest adding dimensions based on "liquidity" and/or "milestones" to improve the most common token lock-up models in the current market.

Liquidity-based Lock-up Mechanism

The definition of liquidity is not absolute, and there are many ways to quantify liquidity in different dimensions.

One feasible standard for measuring liquidity is to examine the buy order depth of tokens on-chain and on centralized exchanges (CEX). By calculating the cumulative sum of all buy order depths, we can obtain a number, which we can call bLiquidity (buyer liquidity) here.

When designing lock-up terms, the project can introduce two new parameters, bLiquidity and p bLiquidity (i.e., the percentage of buyer liquidity, theoretically any value between zero and one), at the contract level, which can be output as:

min (tokens to be claimed under normal vesting output, p bLiquidity * bLiquidity * token unit FDV)

Next, we will explain the operation of the liquidity-based lock-up mechanism through an example.

Suppose the total supply of a token is 100, of which 12% (12 tokens) will be allocated to contributors with lock-up requirements, and the price of each token is $1 (for simplification, assume the token price remains unchanged).

If a time-based lock-up method is adopted, assuming these tokens will be linearly released within 12 months after TGE, this means contributors can unlock 1 token per month, i.e., $1.

If additional lock-up terms based on liquidity are added, assuming the p bLiquidity value set in this lock-up term is 20%, and bLiquidity is $10 (i.e., the token has at least $10 of buyer liquidity within 12 months). In the first month of lock-up, the contract will automatically check the $10 bLiquidity and multiply it by the 20% p bLiquidity value, resulting in $2.

According to the min function provided above (taking the minimum value between the classic mechanism and the additional mechanism), the contract will automatically release 1 token at this point because the release value according to the classic mechanism ($1) is less than the release value according to the additional mechanism ($2). However, if we change the bLiquidity parameter mentioned above to $2, then the contract will automatically release 0.4 tokens because the release value according to the classic mechanism ($1) is greater than the release value according to the additional mechanism (20% * $2 = $0.4).

This is a potential way to dynamically adjust the lock-up structure based on liquidity.

Advantages

  • The mainstream lock-up models in the current market mainly focus on the time dimension and may indirectly consider whether there is sufficient liquidity at specific prices to absorb unlocking. The liquidity-based lock-up model requires the project to actively focus on building liquidity around its token and integrate it with specific incentive measures.
  • For investors in the open market, they will also gain stronger confidence — only when there is sufficient liquidity will the scheduled token amount be unlocked, otherwise only a portion of the scheduled amount that matches the liquidity situation will be unlocked, thereby avoiding a sharp drop in token price due to insufficient liquidity to accommodate new selling pressure.

Potential Challenges

  • If the token consistently fails to obtain sufficient liquidity support, this may significantly prolong the period for contributors to receive rewards (unlocking).
  • Additional rules may complicate the calculation of unlocking frequency and period.
  • It may incentivize false buyer liquidity. However, this can be mitigated through various methods, such as considering only a certain proportion of bLiquidity near the current price, or considering LP positions with certain lock-up restrictions.
  • Contributors may continuously obtain tokens from the unlocking contract but not immediately sell them, gradually accumulating a large amount, and then sell all the tokens at once, which may have a significant impact on liquidity and lead to a drop in token price. However, this situation is similar to whales actively accumulating a large amount of tokens in circulation, and the risk of whales liquidating and causing price drops always exists in the market.
  • It is more difficult to obtain the bLiquidity value within CEX compared to obtaining it within DEX.

Before continuing to discuss milestone-based lock-up models, the project should consider how to attract sufficient liquidity to ensure "normal" unlocking progress. One potential approach is to reward locked LP positions through incentive measures, and another approach is to find ways to attract more liquidity providers — as we wrote in "10 Things to Consider When Preparing for Your Token Generation Event (TGE)," allowing liquidity providers to borrow tokens from the project's inventory to attract more participation and create a more stable market around their own tokens.

Milestone-based Lock-up Mechanism

Another additional dimension that could improve token lock-up models is "milestones," such as user count, trading volume, protocol revenue, total value locked (TVL), and other quantifiable data parameters that can be used to evaluate the attractiveness of the protocol.

Similar to the liquidity-based lock-up design mentioned earlier, the protocol can also design a binary token lock-up clause by introducing additional parameters for each milestone.

For example, to achieve 100% "normal" unlocking, the protocol must reach a $100 million TVL, 100+ daily active users, and over $10 million in daily average trading volume, etc. If these values are not met, the final unlocked token amount will be lower than the initially set target.

Advantages

  • Milestone-based lock-up mechanisms ensure that when a large number of tokens start to unlock, the protocol will have a certain level of attractiveness and liquidity.
  • Less reliance on the time dimension.

Disadvantages/Challenges

  • Data can be manipulated, especially statistics like active users and trading volume are more likely to be manipulated. On the other hand, the TVL metric may be less susceptible to manipulation, but its importance is relatively low for certain projects with heavy capital efficiency. Income is more difficult to manipulate, but certain activities (such as wash trading) can generate more fees, thus indirectly making it manipulable.
  • When assessing the possibility of data manipulation, it is most important to consider the motivations of various groups. Teams and investors (the groups involved in the unlocking plan) have motivations to manipulate statistical data, while investors in the open market are less likely to manipulate statistical data because they have almost no reason to accelerate unlocking.
  • Off-chain legal agreements may greatly mitigate the malicious intentions of groups with such motivations. For example, projects can establish severe penalties for rule violations in advance — if team members or investors are found to engage in wash trading or other data falsification, they may be deprived of their original token shares.

Conclusion

The current market trend of "high FDV, low circulation" has raised concerns among investors in the open market about the sustainable investment potential of the market.

Traditional lock-up models based solely on the time dimension cannot match the complex market environment. By integrating dimensions such as liquidity and milestones into token lock-up terms, projects can better align incentives, ensure sufficient depth, and ensure the attractiveness of the protocol.

Although these new designs also bring new challenges, a more robust lock-up mechanism clearly brings more benefits. Through careful design, these improved lock-up models can effectively increase market confidence and create a more sustainable ecosystem for all stakeholders.

免责声明:本文章仅代表作者个人观点,不代表本平台的立场和观点。本文章仅供信息分享,不构成对任何人的任何投资建议。用户与作者之间的任何争议,与本平台无关。如网页中刊载的文章或图片涉及侵权,请提供相关的权利证明和身份证明发送邮件到support@aicoin.com,本平台相关工作人员将会进行核查。

Share To
APP

X

Telegram

Facebook

Reddit

CopyLink