Skip to content

lawofambak-research/Decentralized-Credit-Ratings

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 

Repository files navigation

Decentralized Credit Ratings

Overview of DeFi

DeFi has allowed developers to create innovative financial products that anybody in the world can use with the use of cryptocurrencies. The current DeFi ecosystem is dominated by DEXs, lending & borrowing platforms, and staking platforms. These services are constantly being worked on to allow users to truly tap into the power of DeFi. However, I believe that one of these areas in particularly, lending & borrowing, will legitimize the use of DeFi and attract a wider audience to further global adoption.

Lending and Borrowing in DeFi

Lending and borrowing in DeFi is one of the earliest use cases of dApps. Briefly speaking, DeFi lending and borrowing is facilitated through various protocols where tokens are pooled together in smart contracts and users can borrow against those pools. This allows for the smart contracts to act as trustless intermediaries instead of centralized financial institutions facilitating these services.

On the lending side, users can deposit tokens into different pools and receive special tokens in return that represent their total value share of the pool. These special tokens can be redeemed at any time for the user’s deposit plus any accrued interest. The interest rates are variable in that they change depending on the token pool size and token utilization rate (amount of tokens that are being borrowed). This means that users who supplied their tokens to a certain platform can withdraw their deposit anytime.

On the borrowing side, users can borrow tokens through over collateralized loans. Due to crypto assets being volatile and riskier than traditional assets, most loans are over collateralized. This means that if a user wants to take out a loan, the user must deposit collateral whose value is greater than the value of the loan. If the value of the collateral falls below the value of the loan, then the user is in the possibility to be liquidated to avoid financial losses. This means that borrowers can acquire loans with no specific payment deadline if the value of their collateral is greater than the value of their loan.

Uncollateralized Loans

Despite the use cases of over collateralized loans, I believe the true power of decentralized lending comes with uncollateralized loans. Just like in traditional finance, most people take out uncollateralized loans such as student loans. Uncollateralized loans will attract a wider audience due to its various use cases. However, the idea of uncollateralized loans in DeFi is not something that simple and easy to implement. In a decentralized environment where lenders and borrowers do not know each other, lenders would need to be willing to provide a loan despite it not being over collateralized. The amount of risk calculation and trust will both need to be considered significantly when creating such a service. To begin implementing such a service, a reputation system between lenders and borrowers would need to be created to establish trust in a pseudonymous market.

Assessment of Credit Risk / Default

Establishing trust between lenders and borrowers in a decentralized environment can be implemented in a similar way as in traditional finance: credit ratings. Just as normal people establish “good credit”, crypto borrowers can also establish “good credit”. However, before talking about the ways this can be done in DeFi, we would first need to understand how it is done in modern day society.

In today’s global economy, there are various methods to determine a person’s credit worthiness. Different countries have different credit rating systems so there is no a “one size fits all” solution. One way to assess “good credit” would be to incentivize good spending, repayment of loans, and no missed payments while penalizing bad credit behavior such as defaults. Another way to assess “good credit” would be to essentially stay off the “bad credit blacklist”. This would mean that people would start off with a certain credit score and would get penalized after every default or missed payment. However, I believe the latter is more vulnerable to risks as any person could create multiple accounts if they ever got penalized on one account. Additionally, this method does not create an incentive for people to borrow responsibly and build good credit. On the other hand, incentivizing good credit behavior and penalizing bad credit behavior is like how credit ratings work in the US and many other countries today. This creates more of a demand for borrowing and is one of the many ways to evaluate borrower risk.

As mentioned above, credit ratings can be created to assess a borrower’s default risk even in a decentralized environment. Just like how a credit profile in traditional finance would include various factors (payment history, length of credit history, and amounts owed), a credit profile in DeFi would include similar factors (transactional data, payment history, and account balance). Although not all factors may be the exact same, I would argue that a good amount of important data are available to be analyzed from the blockchain. Things like payment history, account balance, and length of credit history can be tracked for a specific address. For example, to calculate length of credit history, one would need to track when a user first initiated a loan and compare that with when the borrower is applying for another loan using block.timestamp. The difference between these values can serve as a measure for how long a user has had a credit history. Overall, a person’s credit profile can be used to calculate a credit score that a borrower can possess. This can allow lenders to determine borrowers’ default risks more accurately and can also allow protocols to determine the type of borrowers in the market.

Example Case

An example of a credit profile could be depicted through a struct in solidity. This struct can hold various data types that can represent different factors that add up to a credit rating/score. As mentioned before, these different factors can carry different weights which can eventually add up to a credit score just like how a FICO credit score is calculated in traditional finance.

This simple test example (for over collateralized loans) shows that when user defaults on a loan, the user’s borrow amount (known as collateral factor from Compound Finance) decreases. Additionally, the default collateral factor is the default amount that a user can borrow against an asset. Depending on whether a user defaults or not, the user’s borrow amount can possibly change if they show “bad credit behavior”. Something to note here is that changing the borrowing amount based on the user’s amount of defaults does not take into account the deposit amount (amount of collateral).

Screen Shot 2022-05-24 at 1 25 01 PM

Although this example test case references Compound Finance and is for over collateralized loans, I believe that a general concept of decentralized credit ratings can be implemented to various DeFi protocols. By creating a general credit profile for each user, each user will have details that reveal what kind of a borrower they are to lenders. Overall, decentralized credit ratings can provide important information in a decentralized lending and borrowing market that can possibly lead to the creation of uncollateralized loans.

Possible Concerns / Comments

Credit ratings and scores in traditional finance may contain private information that users may want to not share. However, in decentralized markets these details only pertain to a certain address without any personal information attached. Additionally, decentralized credit profiles can contain different information as opposed to traditional credit profiles (borrower default risk can be assessed in different ways using different measures).

Credit ratings should be compatible with any DeFi lending and borrowing platform. If only one DeFi lending and borrowing platform was to offer decentralized credit ratings, then when those users were to use another platform, they would essentially start off with a clean slate. No matter what platform users use, their credit ratings should follow them so any platform can determine the different types of borrowers.

Decentralized credit ratings should not be the only tool to consider when thinking about the implementation of uncollateralized loans. To allow for a safe place for lenders and borrowers to transact with each other in a decentralized way, financial risks should be calculated through various lenses. In a market where users do not know each other, DeFi protocols should create an environment where users do not behave badly in their own interests.

About

Discussion about decentralized credit ratings

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published