Marginfi token mechanics

Hey guys - great to see everybody here!

Overall note: It’d be helpful to keep this discussion thread around tokenomics and token mechanics vs. airdrop criteria. There’s another thread to discuss airdrop criteria - let’s keep that conversation over there.

In general - @Albert might be helpful to change this topic & airdrop criteria to “Token mechanics” and “Airdrop” respectively, so that we could bucket conversations more effectively.

Related to tokenomics & mechanics, following up on @Albert 's earlier points, I think it’s productive to continue the conversation based off of the following assumptions:

  • The token is meant to be a single governance token for every product, e.g. lending/LST/yielding stablecoin
  • The focus is to operate with a DAO structure and let the community slowly take over the protocol’s management

This can be framed simpler. Let’s assume:

  • There’s one token
  • A decentralized governance system needs to be put in place.

I’m not saying anything innovative so far - this is the standard baseline for a decentralized, token-driven protocol.


On to some more specifics:

I like the idea of governance being done through a staked $vMRGN token. It comes with the following benefits:

  • Time-dependent lockups mitigate voting manipulation
  • More broadly, it requires intention - token holders interested in actively being involved in voting need to lock up stake to distinguish themselves from passive holders with no interest in governance.

The abstract idea of a lockup mechanism is helpful, but falls into the ‘generally good practice’ bucket for me. Assuming all this, so far we have the following three points:

  • There’s one token (let’s assume: $MRGN)
  • A decentralized governance system needs to be put in place.
  • Governance is done through $vMRGN tokens, which you get by locking up $MRGN for a minimum time period.

Okay, here’s where it gets interesting: I think a big problem to solve for is the immediate halt to development progress that DAOs have gotten infamous for. If you look at historical governance systems outside of crypto, this is where the evolution to republic systems comes in - decision-making can be delegated to members voted in, so that voters aren’t asked to participate in every single decision the DAO makes. there’s lots of dimensionality to this (you can structure republics many different ways) but I think it’s overall important that the long-term velocity health of the protocol is prioritized.


Internally, core has been exploring an iterative futarchy approach to governance that I’d like to get folks’ thoughts on.

Here’s a proposed structure under an iterative futarchy model:

  1. vMRGN for Participating in Governance Markets: Holders of vMRGN, which you get by locking up $MRGN, can participate in governance prediction markets. These markets will focus on key performance indicators (KPIs) of the DAO’s actions or potential decisions.

  2. dMRGN for Proposing Governance Markets: To propose governance markets, holders need to lock their tokens for a longer period, e.g. 6 months. This ensures that proposers are committed to the long-term health of the protocol. These dMRGN holders can propose actions or policies, each with associated KPIs.

  3. Governance Markets Functionality: In these markets, holders can bet on the outcomes of various proposed actions. For example, if a proposal is made to integrate a new feature, market participants can speculate on whether this will positively impact a specific KPI by a certain date. This could include metrics like user adoption, transaction volume, or other relevant measures.

  4. Iterative Feedback Loop: The outcomes of these governance bets then feed back into the decision-making process. If a majority of vMRGN holders bet that a particular action will have a positive outcome, the DAO may choose to implement that action. This is the ‘iterative’ part of the process - decisions are continuously informed by market feedback.

Other general things:

  1. Revenue and Rewards: Revenue generated by the DAO (from fees, liquidation premiums, etc.) can be partially used to fund the DAO’s operations and growth. A portion of the revenue might also be allocated to reward active participants in the governance markets, aligning incentives for healthy participation.

  2. Annual DAO Wellness Vote: An annual wellness vote can be conducted, where all vMRGN holders can give their feedback on the overall performance of the DAO. This can be seen as a holistic check on the health and direction of the protocol.

  3. Dynamic Adjustments: The DAO should retain the ability to adjust the parameters of the governance system (like lockup periods, reward structures) to ensure it remains effective and fair.

In essence, this model uses market mechanisms to synthesize information and forecast the outcomes of governance decisions. It allows for a more dynamic and responsive governance structure, where decisions are continuously informed by the community’s predictions and feedback. This should, in theory, help mitigate the slowdown in development progress often seen in traditional DAOs.

Of course, this is just a preliminary outline. I’d love to hear everyone’s thoughts, especially on potential risks, improvements, and the practical aspects of implementing such a system.

6 Likes