Author’s note: this article continues the introductory series by Republic and Byzantine Solutions, conveying vital information an investor needs to know about decentralized finance. The series covers individual instrument groups, supplementing the recently issued report Redefine 2020: A Primer. This article (and the rest of the series) is supplemental reading for those that have attended our Blockchain 101 and Intro to Defi webinar.
The gap between the crypto and off-chain worlds is narrowing, but there’s still some distance left. Purchasing crypto and using it to buy goods and services is still challenging for many people. The factors vary, but typical roadblocks that hamper access to crypto involve factors like where someone lives, their relationships with financial institutions, and their wealth, to name a few.
While steps have been taken to better streamline these processes, there is still a lot of work to be done. Thus, the ultimate fate of cryptocurrencies and decentralized blockchain-based solutions remains somewhat cloudy.
This uncertainty brings to light several questions
Is DeFi on a quest to replace traditional finance?
Can DeFi carve out its own niche in traditional finance?
Could it potentially create new multi-trillion dollar markets that lie between Main Street and Wall Street?
If scalability and interoperability—the technological weak points of blockchain in general—can be solved in the next few years, should decentralized stablecoins ultimately replace the current wire transfer ecosystem? The best way to answer questions like these is to look at use cases that highlight relevant business transactions across the divide between the two worlds.
One such example to point to is the massive amount of crypto capital that has been poured into the industry lately. Currently at approximately $1.7 trillion globally, these underserved, off-chain money markets are providing crypto holders (on chain) with the opportunity to lend off-chain.
Loans and Repayments
The current growth cycle around DeFi gained most of its momentum with the introduction of liquidity farming, which involves decentralized protocols paying token incentives for liquidity provision. The demand side for this liquidity generally falls into one of two categories:
Collateralized lending protocols (which we discussed recently)
On-chain trading protocols (which will be covered in our next article)
Both categories offer a decentralized guarantee of repayment for liquidity providers. A blockchain-based automated market maker has a similar function, but has to rely on excess collateralization and threats of liquidation to enforce repayments from borrowers.
This dynamic creates a vicious cycle of on-chain lending: to get a loan, one needs to put up collateral worth more than the amount of the loan. On top of that, only tokenized assets are accepted.
This doesn't work very well outside of financial markets. One of the biggest use cases for loans in off-chain, brick-and-mortar business is covering cash shortages and funding short-term investments into trade goods or marketing efforts. Trade goods may turn over slowly or lose value while they sit on the shelf, and marketing expenditures might not pay off. These risks make excess collateralization an unappealing proposition.
At scale, rates of default (if they are known and stable) are offset by cost of money, but that takes extensive knowledge of the underlying business and may warrant additional hedging. In a decentralized setting, neither is conveniently available. But at the same time, the protocol must be able to repay its liquidity providers, otherwise nothing works.
To summarize, the main problem with decentralized loans is how to process a default without introducing easily gamable mechanisms into the system. As we discussed, using excess collateral is a natural approach—and it works rather well—but still doesn’t close the gap to the off-chain world.
The two biggest lending platforms on Ethereum, Compound and AAVE, hold deposits of about $10.8 billion in crypto assets. Many view these protocols as safe havens (aside from the potential technological and protocol failure risks) that provide borrowing without collateral.
A natural extension to the protocol, implemented by AAVE v2 in the summer of 2020, was called credit delegation. It allows a depositor to deposit funds in the protocol to earn interest and delegate borrowing power (i.e. their credit) to other users. The enforcement of the loan and its terms are agreed upon between the depositor and borrowers, which can be either off-chain via legal agreements or on-chain via smart contracts.
The depositor (delegator) to earn extra yield on top of the yield they already earn from the protocol,
The borrowers (delegatees) to access an uncollateralized loan.
This is a somewhat niche solution, since the depositor is entirely at the mercy of the party they delegate credit to. Yet, it has its benefits, such as being an anonymous depositor functioning to, for example, fund DeversiFy’s market making.
Credit fund and off-ramp
An alternative approach would be to introduce one trusted third-party to deploy capital to off-chain lending organizations. The trusted party vets a limited number of partners, uploads as much data about them as possible, and then offers it up to the scrutiny of the decentralized community.
The community funds are then pooled to a location which in turn extends credit lines to these lending organizations. Multiple configurations would need to exist to accomplish this. For instance, the central party may have to take junior debt on these credit lines, or credit line allocation may be governed by a DAO (decentralized autonomous organizations, etc. This approach, currently undertaken by Goldfinch, has a certain degree of centralization that may or may not be inevitable. It remains to be seen whether it can (or should) scale up from a largely centralized stage to something else.
Credit scoring as a prediction market (of sorts)
This approach attacks the credit scoring side of dealing with defaults. Prediction markets by themselves are a back-and-forth topic, both in and out of crypto but the gist is that the market should be able to estimate the possibility of a particular outcome better than any central agent. One such construction has been proposed by TrueFi, but appears to be very experimental at the moment.
Lenders first place liquidity into the community pool. Then potential lenders make loan requests to the DAO, while the governance token holders take positions according to their estimation of the possibility of default. They act as traders on the prediction market for the event of the default. If a certain level of confidence is reached, the loan is distributed from the community pool.
Depending on the outcome, people that predicted correctly are rewarded, while those that were incorrect are penalized. The Foundation may then take centralized off-chain legal action against the defaulting borrower. A potential tweak to that description could be an attempt of recapitalization of the community fund by penalizing the governance token holders, but implementing this mechanism would further complicate the design of the protocol.
As new mediators are introduced into the system— which is a usual go-to measure to offset the level of complexity present from overcollateralized lending—volatility risks tend to increase. In addition to this, it is also easier to operate in stable coins, since the borrower operates in fiat, (which we covered recently) and target fixed rates.
With the exception of credit delegation, undercollateralized loans remain a very experimental topic, even when compared to DeFi itself. It is very easy for a centralization bottleneck to occur, or the over-delegation of power leading to vulnerability to hackers. On the other hand, a successful design with checks and balances, and permissionless market-driven arbitrage can position DeFi to supplement both the financial ecosystem and the brick-and-mortar business world without trying to replace either one entirely.