Thread by @ZeMariaMacedo

Mars isn't just another money market

It’s a generalised credit protocol enabling both peer-to-smart contract AND smart contract-to-smart contract lending 🔴

The latter is the real breakthrough and what I’m most excited about.

Thread about why 👇

https://twitter.com/mars_protocol/status/1489262818389946371

1/ Many say DeFi is inefficient because all lending must be overcollateralised – you have to lock up more than what you borrow

They then argue the solution is moving to some form of trust-based "uncollateralised" lending

We agree with the problem, but not with the solution

2/ Rather than porting concepts directly from TradFi, we believe in using crypto’s inherent advantages to create new, better primitives

In order to understand these advantages, we must first examine how credit works in TradFi and how this differs from crypto

3/ In TradFi, debt is a legal construct

Specifically, lenders lend money to borrowers who promise to pay back principal + interest over some time period

If they fail to do so, the lender has legally protected rights to enforcement

4/ Therefore, debt is nothing but a legally enforceable promise. It cannot exist without legal agreement and cannot be enforced without courts of law

5/ Enforcement in TradFi is notoriously expensive, with liquidation involving messy, time-consuming legal processes.

As a result, much more time (and money) must be spent establishing trust in counterparties via cumbersome KYC, credit & background checks, business plan assessment, etc

High cost of enforcement → high cost of establishing trust → higher interest rates

5/ Crypto replaces legal enforcement with systems of incentives. This reduces the cost of enforcement and thus eliminates the need to establish trust

There are two elements which combine to make crypto lending different:

(1) Smart contract lending
(2) Tokenised collateral

6/ SC lending

In the real world, lenders lend to counterparties and must trust them to use the money the way they said they would

As a result, in TradFi much more time (and money) must be spent establishing trust

7/ In crypto, lenders can lend to SCs where they only need to examine the code to figure out what it’s doing and whether they’re willing to underwrite the risks

8/ Tokenised collateral IRL, borrowers may hold all sorts of collateral.In order to liquidate, lenders must go through expensive and time-consuming legal processes with no guarantee of getting their money back (e.g. borrower flees the country or declares bankruptcy)

9/ In crypto, SCs can hold tokenised collateral representing a borrower’s assets. As long as the tokens are liquid and the SC & game theory function as designed, the lender can underwrite knowing a defaulting borrower will be liquidated as soon as his margin threshold is breached

10/Assuming collateral is liquid, all the lender must do is ensure borrowing SC includes appropriate risk params and liquidation logic

Once the borrower reaches their liquidation threshold, anyone can come in and liquidate their collateral, receiving a bonus for their service

11/ Crucially, the loan is trustless in that the lender doesn’t need to know anything about the borrower in order to make a decision (no more credit history), nor do they need to rely on the legal system to make sure they get paid in case of default

12/ To make this less abstract, let’s look at an example of it in action such as the Leveraged Yield Farming (LYF) initially pioneered by @AlphaFinanceLab

LYF is an example of SC lending

13/ In this case, the entrepreneur is the farmer. He’s bullish ASTRO and wants to accumulate as much of it as possible through farming the ASTRO/UST pool

Normally, he’d be limited by the amount of spot ASTRO and UST he holds

14/ A traditional money market wouldn’t help since he can only borrow less than what he puts in

A traditional lender would never lend to a farmer, but if they did they’d require a business plan, KYC and a high interest rate to compensate for the tx costs and risk of default

15/ Instead, the entrepreneur can apply for a credit line from the Red Bank. In order to do so, he writes a smart contract which does the following:

15/ The entrepreneur is now levered 2x long, earning 2x the yield. But isn't this risky for the lender?

No, because the SC holds the LP shares as collateral and enforces a certain margin requirement. If this level is reached, anyone can call a function which does the following:

16/ Let’s put some numbers on it to see how it works

An entrepreneur uses 500ASTRO worth $1 each to borrow $500UST. His collateral is $1000 of ASTRO-UST LP shares and he owes $5000 UST.

His LTV is 0.5 and the SC enforces a margin requirement of 0.83

Over time, as long as the yield on the collateral is greater than the cost of the debt, the user makes money

17/ However, if the value of $ASTRO drops to $0.35 the user now has only 259.8 UST and 845.15 ASTRO for a total of $591.61

The LTV goes to 0.845 and the liquidation condition is triggered. What happens now?

18/ While Mars starts out focused on LYF, this is only the first of many potential use cases

The same architecture could for example be used to extend credit lines for leveraged trading

19/ We can also imagine a decentralised FTX like product on top of Mars, combining LYF and margin trading into 1 credit account with a single margin requirement

The Martian council would just have to underwrite the SC risk engine and liquidation conditions for the various assets

20/ Note that the same design can be used to give arbitrary amounts of leverage with the only limit being the contract’s ability to liquidate which itself is determined by:

a) the speed of the blockchain
b) the liquidity, volatility and other risks of the collateral asset

21/ For instance, many mAssets are a few standard deviations less volatile than most cryptoassets, potentially enabling much higher leverage on farming/trading

22/ Inspired by @AaveAave's pioneering work with their asset listing risk framework, Delphi Labs will release an open-source framework to help the Martian council assess LYF strategy risks, including recommended risk parameters for different kinds of assets and strategies

23/ This is a starting point and will be discussed, iterated and improved on by the Martian Council over time

We envision a future where a smart contract can be built around any kind of liquid, tokenised collateral, unlocking efficient leverage on anything that can be tokenised

24/ For instance, once in-game assets are tokenised and liquid, someone could write a smart contract that uses in-game assets as collateral to borrow stablecoins and buy more of these assets and “farm” with them (i.e. competing or breeding)

25/ The lending contract would allow them to interact with whitelisted contracts (i.e. game contracts) and track the value of collateral and cost of debt over time, imposing a certain margin requirement at which point it’d seize and liquidate the collateral

26/ As more assets and cashflows move on-chain, more and more of the world’s capital will be able to be tokenised and programmed

This will dramatically expand the scope of the kinds of smart contracts that can receive credit lines from Mars

27/ I truly believe the sky is the limit for what can be done with such a credit protocol primitive, and look forward to seeing where the Mars community decides to take it

Disclaimer: I hold $MARS tokens and so does Delphi Labs

28/ Disclaimer 2: All this stuff is experimental technology, built on top of more experimental technology. Please exercise caution when interacting with it and don’t put in more than you can afford to lose. Full disclaimers are available below

mars-protocol.medium.com/mars-disclaime…