The second deliverable for the CoLab Fellowship is the visual representation of the findings.

**Two maps were created to facilitate the understanding of the findings:**The first map shows the problem from the view of the IRL Borrower and the Web3 Lender, separately. The second map details the criteria required to build a solution that effectively bridges web3 lending to IRL micro businesses. Furthermore, this second map unfolds the incentives, constraints, service providers, and potential collaborators that combined could properly solve the real world problem of fair credit access through the use of the blockchain technology.

This 2nd deliverable focuses only on the maps, but stay tuned because the last deliverable is the complete article with market sizing, detail on the solution criteria, recommendations for builders, and a ton of bibliography used for this research.

This post has the following structure:

1. TL;DR

2. Incentives and Constraints

3. Potential Collaborators

4. Conclusion

TL;DR

In this 5 min video the two maps are explored:

Map 1. Players and their context

Map 2. The criteria for a potential solution

Incentives and constraints

The clearest incentive comes from bringing international interest rates to local businesses via DeFi. A local business in México is used to pay 365% a year whereas the web3 lenders are used to get a return of about 10% a year from fixed or dynamic lending protocols.

A progressively lower interest rate could be offered to local businesses in México while a higher lending yield would be possible for web3 lenders.

However, that’s not all. The blockchain industry aims to bring onboard more users. In a bull market there is a spring of new blockchains with dozens of grants to reach out to new users, but new adopters of web3 cannot enter DeFi unless they have a positive credit history, and in its absence, a collateral to borrow.A delegated collateral function, introduced by Aave (Kulechov, 2020) allows others to put up the collateral for someone else to borrow. Although a great function, it has not yet been widely used.

A protocol or a chain could be incentivized to put the collateral so users join their network and ecosystem of dApps, and in the process build a decentralized credit score to enable these new users to use DeFi without depending anymore on a third party support.

While the delegated collateral function exists, and there is the goal of blockchains and DeFi protocols to onboard new users, there’s still something missing: an initial web of trust.

Findings from interviewing web3 lenders showed that a valid web of trust can be initiated by a third party with knowledge, infrastructure, and personnel on the ground near the customers, with the goal to identify the debtors and enforce payback.

This third party is incentivized to provide the service of identifying, onboarding, allocating credit lines, and enforcing payback for a fee or for a portion of the interest yield. As well as, offering the service of facilitating the payback in crypto through an on and off ramping service, in a p2p fashion.

However, when the interviewees were asked about what would make them trust this on-the-ground company, they answered:

a) A smart contract-based penalization that could be applied to a staked amount if the company acted wrongly or poorly;

and/or

b) Sharing co-ownership of such company, so to have a stake on the ways the company operates.

During the interviews, web3 lenders reported they are interested in bringing adoption of blockchain and that they would increase their lending amount once the borrower has build his or her on chain data.

What about the constraints?

Onboarding local borrowers to DeFi means tokenizing data, otherwise centralized control is just turned from the meat and bone realm to the digital one. Tokenizing real world data into a trust score requires not only the technical capacity to do so, which is currently solvable by using Ethereum Attestation Service and Reclaim Protocol, for example; but the knowledge and test of scoring different attributes from a single source of data to show the weakness or strength of it. An example:

Freddy has his corner store in Google Maps, but so his competitor on the next corner. How would one assess the risk of lending to each of them when resources are limited?

Both businesses can differentiate between each other based on the number of reviews they have and the score they got. The 4.6 stars from Freddy sets him apart from his competitor that has 3.2 stars. But that’s not all. The quality of the reviews matter. Even if both businesses had the same star scoring, if one business has reviews from new accounts or from bot-like profiles, and the other has reviews from “local guides”, it makes a universe of difference.

There are other factors within that same data source that make a difference, such as if one business has pictures and the other not, and the flow of people that business receives. All of these data points can be taken from Google Maps. But, GMaps is not the only source, there are plenty more.

This raises several new questions such as:

• Which data points from a data source are relevant to build an initial trust score?

• How much trust value should each of this scoring elements have?

• How could a person create their profile, atestate their data in a verifiable way while preserving the individuals privacy rights?

• How could the attestation be scaled and decentralized?

Potential collaborators

• Ethereum Attestation Service for attesting the trust score and decentralized credit score;

• Reclaim Protocol for bringing web2 data on chain in a privacy-preserving way with zK proofs;

• Token Engineering Commons and the Regen Score initiative from the Trusted Seed could be allies in defining the right scoring mechanism and criteria;

• Arbitrum could be a potential interested stakeholder via a growth experiment where funds could be used to secure under-collateral loans;

• The main players within the DeFi ecosystem could be interested in providing the capital influx for lending;

• DAOified version of an on-the-ground micro financial company that commits to identify, onboard, allocate, and enforce payback of credits, as well as to facilitate both on and off ramping options.

Conclusion

DeFi can bring fairness to the way lending and borrowing works in the real world. To achieve this web3 lenders require an initial mechanism to trust if the person at the other side of the screen will pay back and that will use his or her money in ways that do not affect the lender.

Certain off chain data sources are considered valuable enough to make web3 lenders risk their money. Web3 lenders reported that they would lend larger amounts once the borrower starts building up an on chain data they can effectively validate. However, it is important to note that web3 lenders do not require accessing the data, but having the complete certainty that the data in regard has been verified.

This post showed the minimum criteria to build a solution that complies with the web3 lenders requirements.

--

If you are interested in the topic, stay tuned! In a couple of weeks the complete research article will come out :)

Follow me in twitter as @HumbertoBesso or @0xUrbanika, and make sure to follow up @rndAO__ for updates on my research peers as well as on opportunities for you to join as a researcher in the next cohort ;)