Hi, I'm Eval

Reading my blog == probably NGMI

Notes on the Terra Bites Podcast about Mars Protocol w/ @ZeMariaMacedo (Head of Delphi Labs)

Published September 6, 2021

Relationship Delphi Digital <> TerraForm Labs

Pete:

“How did you end up building on Terra ecosystem?”

José:

  • Interest in algorithmic stablecoins
  • Terra well developed and under the radar
  • Delphi Labs active contribution = help building missing core primitives (money market, AMMs)
  • Not maxis (also look into Solana), but most efforts focused on Terra at the moment

What is Mars protocol? in comparison to Anchor and money markets like Aave/Compound

  • Mars = on-chain, non custodial, community-governed bank
  • Attracts liquidity and provide lending while managing insolvency risks

Mars compared with Anchor

Anchor:

  • Anchor = savings as a service
  • Solely focused on providing a fixed-rate (~20%) on UST deposits
  • Only accept yield-bearing collateral (PoS assets) in order to achieve fixed savings rate
  • Anchor ~= specialized Mars, closest equivalent to crypto risk-free stable savings account

Mars:

  • Mars = decentralized banking protocol.
  • Rates fluctuate and are determined by the pool(s) utilization rate (borrowing demand)
  • Unlocking leverage on non-yield bearing assets (e.g: BTC)

Mars compared with Aave/Compound

  • Existing Terra assets (e.g: mAssets) will be there as collateral options
  • Onboard assets from other chains with IBC & Wormhole (Col. 5)
  • Three main differences compared to ETH money markets:

1) Uncollateralized lending:

TradFi markets:

  • Loans are capital inefficient (low LTV)
  • Exchanges are custodial and can liquidate you
  • Small addressible markets since only depositors can borrow

Same as if Airbnb would only let hosts stay at each others' houses

Mars design:

  • Allows governance to extend uncollateralized credit lines to smart contracts
  • Extends to a largely untapped market –> non-depositors borrowers
  • Similar to offering of existing ETH protocols like Alpha Homora & Iron Bank
  • Sounds risky but ~= bank extending credits to businesses based on credit history, expansion plans, assess whether they can pay back
  • Mars “only” has to deal with smart contract risk. Audit code, review liquidation logic, assess overall risk level –> decide how much credit it can allow
  • Embrace crypto trust minimization instead of re-creating TradFi in crypto
  • Delphi Labs working on smart contract risk assessment framework
  • Long-term any protocol can open credit line on Mars

Example (leverage while staying long $MIR):

![[mars_mir_example.png]]

2) Token economics

Revenues:

  • Mars generates fees on the borrowing side
  • No performance fees on Mars directly (but Dapps built on Mars might have some)

Governance:

  • In Cream/Compound, governance tokens holders participating in votes do not bear the consequence of their action
  • In Mars, governors (the Martian Council) need to have skin in the game == $XMARS staked
  • $XMARS stakers suffer first in case of bad assets introduced to the system

3) Dynamic interest-rate model

Traditional MM:

  • Set optimal utilization rate depending on asset
  • 100% rate = illiquid, withdrawals impossible
  • Target optimal utilization lower than 100% to stay liquid
  • Manipulate % interest rate to keep it in control
  • Can’t react to external market condition (e.g: people are willing to pay a high interest rate if the %APR is sufficiently high)

Mars / Oiler Network (Ethereum):

  • Block-by-block dynamic system that continuously adjust depending on utilization rate & optimal target rate
  • Based on control theory (active contribution from physics PHD academic)

Solvency issue scenario

Maker & Aave:

  • Maker = mint $MKR to backstop the protocol in case of $DAI insolvency (e.g: March 2020)
  • Aave = diversified safety module backstops the protocol
  • Governors must have skin in the game

Mars:

  • Above solution has a flaw

  • Negative reflexivity in case of shortfall event.

  • Example with AAVE:

    Anticipate dsafetselling pressure from safety module -> people sell AAVE in advance -> safety module has to sell more -> protocol mint more tokens -> increased selling pressure -> repeat

  • Tranche-based system instead:

    Tranche 1 Insurance Funds (aUST) - from protocol-generated fees. First one to be wiped out in case of shortfall event

    Tranche 2 Martian Council (MARS) - protocol governance token holders

    Tranche 3 (community decision) - issue a debt token like Yearn / mint tokens / do something else

![[mars_solvency_tranches.png]]

Launch & future plans

Expected launch:

  • Waiting for Columbus 5 update
  • A month or two from audits (as of early September)
  • No information about token release for now
  • Mars' future will be determined by the community as more assets come into the protocol

Future plans:

  • Co-incubated by Delphi Labs (risk framework), IDEO (design), other contributors (smart contract dev)
  • Big goal = create innovative governance structure to coordinate all these contributors
  • DAOs are the most interest part of crypto but are very innefficient at the moment
  • Inspiration from Yearn governance 2.0

Show notes:

Read the Mars Red Paper - https://mars-protocol.medium.com/mars…