GIP-35 Untangled <> Goldfinch: RWA reporting

  • Authors: Manrui Tang, Untangled Finance

    • Introduction:

      • Untangled Finance develops RWA origination technology for DeFi. We focus on supplying RWA collaterals to DeFi credit and stable coin protocols (see our progress with Maker and Celo)
      • The founding team has
        • Audit, due diligence (financial, commercial) and strategy experience with a Big 4
        • 20+ years of experience working with global financial institutions/institutional investors in both developed and emerging markets
        • Chartered accountancy qualifications (ACA, equivalent to CPA in the US)
        • Deep experience in emerging markets and strategic partnerships with lenders in Africa: Binkabi, our loan origination product, is being used to originate loans by lenders in Africa.
  • Summary:

    • Untangled Finance proposes to provide RWA reporting service to Goldfinch. This will enable a standard, robust and timely (daily) view on performance of loans, their off chain collateral (as applicable) and covenants at loan-level and portfolio-level across all borrowers. We provide professional and independent review and analysis of borrower’s data and covenant compliance to identify risks in time for remedial actions.
  • Motivation:

    • With the crypto market reeling over recent high-profile failures, there is increasing demand for transparency, particularly proof of assets and proof of liabilities (proof of solvency). For RWA protocols there is often an element of centralisation since the collateral assets and underlying legal agreements are off-chain.
    • Goldfinch, as a under-collateralized lending protocol relying on trust, should be at the forefront of transparency with respect to RWAs
    • Institutional investors, the key to Goldfinch’s growth, will demand the standard of disclosure to be at least as robust as in TradFi before deciding to supply liquidity to the protocol (e.g. Aave)
  • Specification & Requirements:

    • Scope of work
      • Create an custom-built analytics system to provide timely insights into the asset performance including both on chain and off chain data:
        • Standardise borrower reporting format to enable comparisons between different asset pools - Continuous
        • Report and monitor of loan covenants
        • Connect with borrowers and their debtors (gradually) via APIs to get most up to date data
        • Tokenize collateral assets on Celo blockchain (as applicable) (low-cost, carbon negative, mobile-friendly blockchain built for the real world, which Goldfinch also has a relationship with)
      • Portfolio reporting and analysis:
        • Key performance indicators based on loan-level data across the entire portfolio - Daily
        • Borrowing base analysis (Daily - frequency to be agreed)
        • Review borrower current reporting and highlight any potential issues and recommend remedial actions - Monthly
        • Onchain Net Asset Value (NAV) calculation using discounted cash flow method and agreed risk parameters (Daily - frequency to be agreed)
        • Simulating bad debt scenarios - Monthly
        • Our deliverables will be rendered via Untangled app for each pool (We will share samples and customise to requirements as we progress with this proposal)
      • On-boarding DD
        • Provide independent and progressional due diligence exercise including industry, originator (borrower), portfolio analysis such as underwriting, origination and servicing strategy, asset performance and collection policy, risk analysis and potential mitigants to a standard expected by TradFi institutional investors (separate scope and fees to be discussed)
    • Fee (for reporting and analysis)
      • 0.15% pa (15 bps) of loans outstanding (currently ~ $150k yearly or $12.5k monthly)
      • Prepaid every quarter - first payment will be ~ $37.5k in USDC
      • After the first quarter trial - contract to be up for renewal annually
  • Benefits:

    • Transparency - increase investors confidence particularly institutional investors
    • Timely reporting on credit and counterparty risks - remedial actions to avoid/minimise credit losses to the protocol
    • Over time, better data and risk monitoring should result in lower return expectation from investors and lower cost for borrowers, including lower interest for end-borrowers in Africa - this is key in attracting institutional investors and high quality borrowers which are crucial for protocol growth.
  • Downside:

    • Need collaboration from borrowers and underlying fintech debtors (where relevant)
    • Reporting and review based partly on data supplied by borrowers - lack of third party verification
  • Voting:

    • Yes: Contract Untangled Finance to provide asset reporting and analysis
    • No: Do not contract
  • Resources:

    • Goldfinch risk management framework
    • Goldfinch risk appetite statement
2 Likes

Hey! Can you add GIP-35: at the beginning?

1 Like

The proposal is good, but the community already voted for similar kind of reporting GIP-20: Financial reporting through Messari’s Protocol Services. I am not sure how much this proposal differs from GIP-20, community definitely don’t want redudant ones.

Hello @mans9841 ,

Untangled proposal differs from Messari proposal in 3 fundamental aspects:

Purpose and audience

  • Untangled focuses on credit portfolio risk monitoring and management. The audience therefore is existing investors/lenders and DAO management teams

Scope

  • We provide in depth credit reporting based on both on chain and off chain data (working with borrowers) with a focus on probability of default (PD) and loss given default (LGD) which in turn drive the Net Asset Value (NAV) calculation of the portfolio. We are not doing analysis on lenders/investors distribution, lender turnover in various pools, protocol treasury or high level borrower analysis. There is very little overlap between the 2 scopes.

Timing

  • We provide data on a much shorter time scale, in any case no later than monthly (instead of semi annually). Again this is because effective credit management requires up-to-date data. Some reports such as NAV calc will be done on a daily basis while others are as and when required.

In a way Untangled service is analogous to that of Gauntlet (but focus on counterparty and credit risk). In the example of Compound, Gauntlet provides risk simulation etc but Messari works on investor relation reports.

1 Like

@mans9841 I do agree with @MUntangled that the two proposals are different - it appears that the Messari proposal focuses on the protocol’s financials, whereas Untangled seems to be focusing on reporting of the pools.

I definitely agree with the spirit of the proposal, along with some of the statements made in the proposal itself:

→ Untangled mentions Aave, but I’d venture to say that any institution with tradfi structured credit roots will demand the same level of reporting as a bare minimum.

However, I do have a few thoughts/concerns:

  • Receiving a fee, by its nature (especially a fee benchmarked as a percentage of loan balances outstanding), makes the reporting not fully independent.
  • Some of the bullet points listed in “Portfolio reporting and analysis” sounds like there will be some sort of analysis done - my personal thought is that if Untangled opines on credit or operational execution, then there may be a deviation from pure reporting (i.e. what trustees for private credit/structured credit deals in tradfi do).
  • Untangled’s core business model is to supply RWA collateral to DeFi, not necessarily white-labelling reporting infrastructure. As much as the RWA DeFi industry has the “rising tide raises all boats” mentality, I would imagine that the credit funds borrowing from Goldfinch and deploying via asset-based structured credit facilities see themselves also as suppliers of RWA collateral to DeFi. I wonder how comfortable the GF pool borrowers will be on working with Untangled.
  • Focusing on PD, LGD, and NAV fall short of what rating agencies do when they report on credit risk. If the intent is to do what rating agencies do, then Untangled would have to calculate PD, LGD at the underlying collateral level, roll that up to the debt facilities, then roll those debt facility cash flows up to the GF funding facilities.

I’m wondering if it makes sense for Untangled to share some examples of reports/reporting (redacted where needed), to provide a better understanding of the work that is being proposed. Otherwise, how would Untangled differentiate itself from other existing tradfi solutions, like for instance, Finley?

Also, just a thought - I wonder if it makes more sense for Untangled to partner with Goldfinch, not to provide reporting infrastructure, but to launch a pool, and provide these reports themselves, to set the standard?

2 Likes

Thanks for the thoughtful comments @JeremyKim. We will start with the more ‘technical’ ones and return to the more ‘commercial’ ones later.

Some of the bullet points listed in “Portfolio reporting and analysis” sounds like there will be some sort of analysis done - my personal thought is that if Untangled opines on credit or operational execution, then there may be a deviation from pure reporting (i.e. what trustees for private credit/structured credit deals in tradfi do).

The above is a stylised collateral flows and proposed RWA reporting at the start and end states.
Fund flows: Goldfinch makes Loan 1 (onchain) to a Credit Fund borrower, secured (offchain) by Loan 2 which the Credit Fund has made to a Fintech Originator. Loan 2 could be secured by Loans 3 to Fintechs. So here Loans 3 is RWA collateral for Loan 2 and Loan 2 is RWA collateral to Loan 1 (Pool loan).

Start: Enhanced pool credit reporting

As we would be bound by existing credit agreements, i.e. if an agreement does not provide for ‘looking through’ underlying collateral performance then the credit fund is not obligated to provide data.
At first, we may not require additional access to underlying originators in order to provide enhanced reporting. We can gather information that indicates their ability to repay the facilities from the credit funds from other sources.

The analysis will be parameterised - for example when we report on NAV we will agree with the community on the parameters such as, PD, LGD, …based on industry ranges for the type of underlying collaterals.

Focusing on PD, LGD, and NAV fall short of what rating agencies do when they report on credit risk. If the intent is to do what rating agencies do, then Untangled would have to calculate PD, LGD at the underlying collateral level, roll that up to the debt facilities, then roll those debt facility cash flows up to the GF funding facilities.

End state: “look-through” on chain credit reporting and analysis

Ultimately the health of Loan 1/Pool Loan depends on the health of the loans/collateral made downstream (Loan 2 and Loans 3). Therefore RWA reporting also needs to extend downstream to collateralise Loan 2 and Loans 3. This look-through reporting, however, requires appropriate legal provisions and reporting capabilities of downstream borrowers. This is the target of RWA reporting at the end state.

As we gather more data we will be able to construct better credit models to predict and scenario-analyze credit performance of pool loans.

Untangled’s core business model is to supply RWA collateral to DeFi, not necessarily white-labelling reporting infrastructure. As much as the RWA DeFi industry has the “rising tide raises all boats” mentality, I would imagine that the credit funds borrowing from Goldfinch and deploying via asset-based structured credit facilities see themselves also as suppliers of RWA collateral to DeFi. I wonder how comfortable the GF pool borrowers will be on working with Untangled.

We are not a credit fund but a tech provider. By supplying RWA collaterals to DeFi we meant any RWA asset owners can use our technology to supply assets to DeFi. We have discussed connecting a credit fund in the Goldfinch ecosystem to asset originators that we sourced.

With this proposal, we are not aiming to replicate TradFi but implement a credit reporting system that is fit-for-purpose for RWA DeFi credit today. This system is based on:

  • Tokenization of RWA collaterals - each collateral is tokenized into an NFT with appropriate metadata
  • Reporting on key credit performance indicators by pools e.g. onchain NAV calc and covenant compliance
  • Credit modelling: inputs into NAV calc with PD, LGD appropriate for the underlying collateral asset class as described above.

Receiving a fee, by its nature (especially a fee benchmarked as a percentage of loan balances outstanding), makes the reporting not fully independent.

Regarding fee we see this is as the protocol outsource the credit risk reporting to Untangled. We propose to be awarded fees from the protocol’s treasury, rather than from originators. This is to ensure that we are independent of the originators who we will report on. This is also inline with TradFi industry practice where fees are paid out of the cash flow waterfall.

As to % of loans outstanding this is as you probably know, an industry practice. Our fee is based on principal outstanding which is an objective number that the protocol tracks.

—-

As to sample reports, we could create a folder and share it with the community.

Finally, we welcome the opportunity to partner with Goldfinch on bridging originators/borrowers to the protocol. We believe we could do this while providing the protocol with credit reporting and analysis with the above specs.

1 Like

Goldfinch is working on a really important solution of DeFi RWA credit - undercollateralized lending - and to get it scaled, as @MUntangled and @JeremyKim pointed out it is important to get institutional investors’ backing. TradFi institutions are already accustomed to the standard of reporting and monitoring (when done right) but DeFi institutions like Maker is now becoming even more rigorous with RWA reporting (gov vote just passed today).

Would love to hear the community’s further feedback on this proposal before deciding to put it forward for a Snapshot vote (or any other governance step, please advise).

1 Like