Manifesto

Blockchain technology allows to build a decentralized on-chain credit solution with real-life use cases.

If credit is a core pillar of the emerging on chain financial system, the evolution towards general adoption of DeFi lending protocols is still very early.

The main challenge lies in the economics: subsets of finance, such as transactions, can be decentralized quite easily, and Bitcoin may be the best example. But dealing with the core of finance, i.e. lending, is much tougher, as it relies on adverse selection model: the lender must estimate the probability of correct behavior from the borrower to get paid back, which is usually done through a proper credit risk analysis.

This analysis is very complex in decentralized and open networks. Although DeFi protocols offer a transparent and accessible financial system, they are based on pseudonymity, which makes credit risk analysis challenging. As a consequence, unsecured loans are nearly impossible in open networks, as an anonymous borrower would have very little incentive to pay back. That's why most current DeFi protocols focus on pawn-lending, and use over-collateralization to mitigate the credit risk: a borrower should deposit a collateral of 150 to be able to borrow 100. In that way, the borrower has a real incentive to pay back, and the system remains viable, even in difficult market conditions. However, the major downside of this approach is that it limits strongly uses cases: a pawn loan system does not allow a user with little capital to acquire an asset.

In traditional finance, most consumer finance products (consumer loans, split payment, etc.) are unsecured, meaning not collateralized by an asset. In decentralized networks, such solutions would be very complex to implement a priori. A solution based on a decentralized identity system, i.e. a decentralized on-chain credit score could work. The latter could integrate historical transactions, past credit performances, and interactions within communities and virtual ecosystems. If this approach could function in the long run, it has two main limitations: (i) a score, which is based on statistics, needs a mainstream adoption to be representative and to measure correctly the risk, and (ii) tracking every on-chain interaction within any web3 project to create a credit score might be questionable. Today's existing unsecured decentralized finance protocols (such as Goldfinch) mostly focus on business loans, for which the credit risk is assessed by the protocol risk team.

At Nemeos, we do believe it is possible to make a user-friendly fully open and censorship-resistant financing protocol allowing anyone to take a loan to buy a digital asset, without having to deposit additional assets in collateral or to rely on an immature decentralized scoring system, under a set of conditions :

• The financing is secured by the acquired asset itself, just like a classic mortgage loan. The asset will be foreclosed and sold in case on non-repayment ;

• The protocol must be decentralized and structured as peer-to-pool approach, to be scalable and offer standardized financing conditions. A financing pool will be made for each collection, allowing LPs to select which assets they're willing to finance ;

• While the credit is on-going, the user should be able to use his assets.