Putting the insurance fund to work

Thanka for putting this together @brian_smith_0, this is a really thoughtful approach.

From my perspective, I don’t think we need to be quite as aggressive putting the treasury to work. Yes, we should definitely be looking to capture some yield, but it doesn’t need to be overly productive, the key thing is to make sure the treasury doesn’t lose value. Everything else is cream on top.

I would suggest a higher % toward uninvested (25%?).

Hey everyone,

as one of the key concerns of the fund is to minimize counterparty risks vs other protocols, I wanted to say a couple of words about Vyper, which is a protocol with the express purpose of effectively managing these risks.

In a nutshell, Vyper allows yield generating deposits to be split in a junior and senior tranche. The junior tranche earns some of the interest of the senior tranche but is impacted first if there is any capital loss (eg due to platform insolvency). This is a simple but powerful concept which is extremely popular in tradFi.
In practice this will be shipped through a first application layer called Vyper Vaults, which will also manage automatically deployment and auto-compounding.

I see protocol treasuries like Mango’s a perfect fit for our “senior tranches” = slightly less yield but with a capital buffer to protect the funds from any capital impacting event. We are also using a modular approach so in the future this will also be combined with e.g. yield aggregators, further decreasing risks.

Though mainnet is still a few months away, I wanted to put it out there and happy to get any feedback in the meanwhile. We participated in the Convergence hackathon and will also participate in Riptide where we’ll integrate the major lending protocols like Solend, Port and Jet (and possibly also Mango deposits too!)


Would it be possible to get more formal discussion on the risk vs. reward of stable lending. What’s the likely smart contract risk involved in the stable lending and what’s the expected return.

1 Like

As an update on this process, there was some discussion about this on the Feb 12 dev call. Starts at 1hour, 25 minute market here: Developer Call - 02/12/22 - YouTube.

Here is the summary sheet referenced: Notion – The all-in-one workspace for your notes, tasks, wikis, and databases..

The takeaways from the discussion are below.

    • Daffy propose 15% SOL and 10% BTC.
    • Mango has SOL expenses from opening markets and running keepers. Just aligning assets with liabilities. Not speculative buys.
    • BTC makes sense as savings vehicle but less attractive since crypto correlation with no liabilities to match. Slightly smaller allocation.
    • Can acquire via Friktion tool to swap with MMs for 100bps spread. Could implement from governance. Friktion has PR now for anyone to build.
    • Next steps:
      • Work on best implementation method.
      • Reach out to Friktion on status of tool and contact market makers for alternative ideas.
      • Put up governance vote on allocation once implementation is possible.
  • Backstop liquidity
    • General agreement it makes sense and is important to reduce systemic risk to Mango of scenario where liquidators are being liquidated. This scenario is extremely rare but important to have liquidity on books when everyone else is getting liquidated.
    • Any occurrence would be very profitable and 5%-plus annual returns expected.
    • Question whether fully automated version is possible or requires someone to run.
    • Next steps:
      • Release parameters for development and associated bounty to community.
  • Offchain Market Maker Loans
    • Structurally possible. Special purpose legal entity can engage in legal agreements with the DAO. Community vote on legal agreement important so people can verify the DAO consent. Some risk from legal action in physical jurisdictions but tail risk.
    • Some concerns about morals of directing capital towards MMs.
    • Next steps:
      • Query market makers for interest and indicative structure.
      • Move forward on creation of vehicle.
  • Onchain lending (directly or via aggregators)
    • Not currently attractive risk reward.
    • Maybe small allocation down the road.
  • Venture funding
    • Have had inbound interest about VC investments multiple times.
    • Strong arguments in principal: aligning interest with partners, access to good projects, Mango can benefit from say in governance decisions.
    • Practically could be messy: distraction from core competency of building best trading vehicle. Legally complex and would require off-chain entities to sign commitment papers (similar to offchain MM loans).
    • Next steps: N/A. Any interested protocols can submit offers to the DAO.

Thanks for providing this @brian_smith_0 ! Really interesting discussion here. Would like to add, as @mschneider pointed out on the call, that the two main costs to think about with respect to backstop liquidity are gas and opportunity cost. However, since it does not seem like the DAO is in a hurry to deploy these funds, the opportunity cost really seems negligible. I think BLP is a great idea and should ultimately improve the health of the protocol.

I also wanted to ask if Mango is considering investing in structured products such as Katana’s Covered Call Vaults to generate yield. I think the SOL Covered Call Vault in particular could offer some solid risk/return if Mango plans to hold SOL tokens to hedge existing liabilities. Is this something worth considering?

Have we ever talked about distributing some of the USDC from the DAO to MNGO holders? Mango Market isnt supposed to be an asset manager and buying BTC and SOL forces the people who want to invest in Mango by buying MNGO to get exposure to cryptos that they dont want to. The DAO is also slow and will not be able to use the capital in the most efficient way as possible.

Voting is now open on proposals for the Treasury to purchase $1mm each of SOL and BTC. Can see the proposals and vote here: https://dao-beta.mango.markets/dao/MNGO.