Taking Lending Off-Chain
On an emerging trend of off-chain lending becoming increasingly dominant
Hey there,
I have been spending time thinking about PleasrDAO's $3.7 million loan from the Ironvault recently. PleasrDAO is an investment collective that owns NFTs worth over $10 million. They took a line of credit against those assets from another DAO named Iron Bank. I will not go into the specifics today, but you can follow this article for details. The premise of the loan itself is pretty simple. You have a DAO with assets in hand that sees an opportunity to buy more assets given market conditions. You get a line of credit against the illiquid asset from an entity looking to generate interest on their idle cash and continue to invest. In some sense, this is a form of leverage. But here's the interesting bit. Since art does not get appraised as frequently as a digital asset like Ethereum, the odds of this debt position being liquidated due to being under-collateralised is relatively low. So long as PleasrDAO can make the interest payments and repay over time, it is a reasonably low-risk transaction.
But not everybody owns NFTs or runs a DAO with millions of dollars in desirable digital assets. Much of the world runs on undercollateralised debt or off-chain assets. The on-chain lending market has north of $20 billion in debt issued currently on layer-1 networks but this is only a fraction of what can be served realistically. For this to scale, three aspects will need to play out
Off-chain assets becoming lending collateral
The intermingling of traditional forms of currency with existing debt related protocols
Improvement of debt recollection processes from the off-chain world
Making off-chain assets a form of collateral helps onboard a new generation of borrowers to DeFi. Issuing debt in emerging markets could be a good idea to make this a reality. Interest rate arbitrage can be done by connecting a world with low-interest yield and ones where capital markets don't service the vast majority of customers yet. This would mean accruing capital from regions with low or negative interest rates (e.g.: Switzerland) and offering it in ones where borrowers pay high-interest rates (e.g., India). What makes this challenging is the variable currency exchange rates and the cost of recollecting debt. If a borrower's debt is denominated in the foreign currency and the local currency slides, they can expect to pay a much higher than the anticipated interest rate. On the flip side, if the debtor chooses to simply default, there are very few means for the lender to recollect their capital, especially in the absence of collateral. So how do we solve this? A new generation of startups in the digital asset space have been trying to do just that. We explore them in today's issue.
Networked Loans For Businesses
The lending market is in the hundreds of trillions of dollars in size and requires different, specialist approaches to capturing it even while the underlying business may be doing similar functions. Much of the present-day activity in off-chain lending is focused on institutions and not individual lending. Part of the reason for this is that they take larger loan sizes, have considerable assets in hand already, and have reputational damage if the loan is not repaid. What does a blockchain help with in this case? In my observation, the contribution DeFi makes is for assisting with loan origination, verification of documentation, tracking changes made for it, and capital movement. To see how this works in practice, consider Centrifuge
Centrifuge
During periods of high crypto volatility, it helps to have exposure to real-world assets. Centrifuge is helping bridge that gap through being a network that can connect DeFi enthusiasts with real-life businesses looking for a line of credit. The flow for a loan origination on Centrifuge is relatively straightforward. Each time a loan is issued, the originator issues a privacy-enabled NFT that captures information regarding the amount, parties involved, region, due date and other variables. The variables can be seen only by the parties taking the loan and funding it, thanks to Zk-snarks. External parties such as an auditor or a valuation firm can plug into the network to offer their services on a need-by-need basis. These NFTs are then passed on to a pool that finances the lending. Stand-alone investors are required to do AML/KYC before they can offer capital to borrowers.
Centrifuge uses a tranched model for assessing risk. A certain amount of protection is offered to users from the case of default through offering two default tranches named Tin (junior, buffer tranch) and Drop (senior, low-risk tranch). Investors buying into the junior tranch stand to receive a higher yield on the investment but will also be on the line in the case of a default. On the other hand, the senior tranch generates yield that is more in line with what a traditional bank can offer. Individuals investing into these debt pools can redeem the amount invested from time to time so long as there is enough capital to enter the system. If I invested today on debt that matures six months from now - I could redeem my capital at month four for an emergency. As long as there are additional sources of capital looking to invest into the debt. (Note: I have intentionally not gone too deep, but if you want to learn the specifics of how this works, the Centrifuge documentation is rich with information).
Centrifuge is a peer to peer debt network that combines infrastructure for debt origination ad tracking with a marketplace for investing and trading them. They are not a stand-alone lending application that interfaces directly with users. Instead, they empower practically anyone from anywhere in the world to verify and list these debt instruments that a global group of investors can invest into using stablecoins. The debt originators are responsible for the repayment of the loan in time by collecting interest and principal from whom they have disbursed loans. This means that so long as the off-ramps and on-ramps are taken care of,* anyone can build a lending application. That counts as democratising finance in my books.
Arboreum
Arboreum is a lending network that is bringing the concept of credit unions to the blockchain era. Small and medium enterprises in emerging markets routinely have payment gaps of 30 to 180 days between an invoice being generated and settled with full payment. This is lost time for a business to scale up operations. This is the gap Arboreum is trying to close today using credit unions. The venture relies on an “anchor lender” - generally a venture already well connected to the borrowers and is willing to provide money as a loan. They give an initial influx of capital, regional know-how and understanding of the business. External investors from DeFi can then lend into a pool that is used to provide debt as a union. The investors receive what is called a “Collateralised Loan Obligation” - an instrument backed by the firm's receivables. These CLOs (yes, they sound similar to CDOs) can be then traded in the free market. Here’s the interesting bit.
Arboreum claims that a decentralised network of nodes and neural networks can do a better job at loan underwriting than a centralised bank with all the rich data it holds. How can this be? Arboreum calls what they offer as infrastructure to form credit unions. Individuals or businesses that working close to a venture can come together and form a credit union to provide capital for part of the loan. The rest of the loan is paid for by an external institutional source of capital or DeFi based pool. Each credit union acts as a node in the network relaying data of repayment frequency and general creditworthiness. If a node issues bad loans, its ability to further finance other lenders will decline as more of the bad debt is passed to the anchor lender. More importantly, it will also affect nodes that work closely with it. The credit scoring used in the system relies on each individual within the node repaying. Reputational and repayment data is stored on-chain and is used to assess what debt ratios look like at a network level. You can see the entirety of their deck here for a more detailed breakdown of how the whole thing works.
Lending To Institutions
There is one more avenue lending ventures have begun looking for yield. Through focusing on financial institutions that are looking to lever up on the markets without liquidating assets. These institutions use the assets in their existing investments and their reputation to receive a line of credit. Maple finance for instance, takes assets from public pools and offers them to verified, renowned institutional clients. They do require backing the loan with assets. However, in the event of a default, the venture does not immediately liquidate the collateral. Users lending assets on the platform receive interest and maple tokens as yield. TrueFi has a similar process of taking public capital and offering it to vetted institutions whose credibility can be verified on-chain. Loans taken by firms are verifiable on-chain and contribute to their on-chain reputation score which is being developed
A gradual evolution of institution-facing lending markets is setting up private, permissioned pools where only known actors can engage with one another. This would ensure that private institutions are aware of the source of capital and counterparties in a closed ecosystem Aave has slowly been trending towards this with their offering named Aave Pro. Once launched, it would allow hedge funds, banks, and other institutions to offer loans to one another in a closed, white-listed environment. Regulatory guidance such as the one issued by the FATF recently could further push DeFi based institutions to this direction. One of the ventures bridging institutional demand for yield is Alkemi.network. The venture handles close to $30 million in TVL from over institutional clients today and was a pioneer in setting up KYC'd, permissioned liquidity pools. Each of these businesses build their edge on a different value offering. Aave has the reputation of being around for the past few years, TrueFi's proprietary credit scoring system may help it separate itself while Alkemi is building on years of expertise and know-how from working on bridging centralized financial applications with DeFi. I highly recommend reading this recently published paper by Consensys for more on how DeFi and centralised financial institutions are intertwining.
What The Future Looks Like
The future of lending without collateral looks bright, but it comes with trade-offs on privacy and decentralisation. Most existing solutions that look at bringing off-chain lending to blockchains create complex financial instruments in the process or have closed-loop lending networks such as the ones offered by Aave Pro. If pseudonymous, undercollateralised lending is the goal - the odds are high, we won't see it soon without much work going towards designing on-chain credit scores. There are few ventures working on this. I will write about them soon. That said, given the amount of idle capital within crypto and money earning little to no yield in the traditional world - it is only a matter of time before we see more founders using blockchains as infrastructure for enabling lending. Centrifuge's $30million+ in loans is proof that there is demand for deploying capital into real-world assets. Between Compound, Aave and other prominent lending protocols - there is roughly $30 billion worth of capital. Barely 1% of that has gone into off-chain lending. A considerable part of what could enable that transition is likely NFTs. If income streams from businesses can be tokenised and traded, we probably see more interest in off-chain lending. A different direction that could take is through income share agreements or financing freelancer invoices. There are endless possibilities.
Related reads from the past
1. A bank for the gig economy
2. Off-chain Credit and DeFi
3. Blockchains in developing economies - Pre-banking Foundations
I will see you guys on Wednesday with a write-up on treasury management. You should join us on our Telegram group if you'd like to discuss some of this.
Peace,
Joel John
__________
Notes.
1. I have direct and indirect commercial relationships with (some) ventures mentioned above
2. Not to be considered investment advice.
3. Call your parents more often.