Can cryptocurrency loans create value without reneging on their original promises?

Blockchain technology, thanks to the security it offers, has enabled the advent of decentralized finance, an alternative to traditional markets. To do this, “smart contracts” are deployed, that is to say computer code governing operations. In this case, the “money” sent must be representable on the blockchain: it is therefore a cryptocurrency. This is, for example, what can be implemented using the Ethereum system.

Decentralized finance aims to reduce the level of intermediation and therefore the costs paid by users. It allows, in principle, to free oneself from the control of centralized authorities as well as the barriers to access to the operations it offers. In addition, smart contract codes are public, as well as the history of operations.

However, this raises several concerns. Will current ecosystems, modest in size, although already significant, have the capacity to take on the scale necessary for widespread and truly useful use? Will regulators have reason to take drastic measures against the protocols, and if so, will they then be able to fulfill their announced innovative function? Questions also arise in terms of security: cyberattacks, scams or computer code flaws can lead to significant losses of funds, sometimes without possible recourse.

Currently, there are two main uses of decentralized finance. The first is automatic market makers whose role is to offer an alternative to centralized currency exchange platforms. The second is that of decentralized lending, the subject of our work with Franck Gabriel, researcher at the University of Lyon I. This has yet to prove its ability to reconcile the ideals of cryptocurrencies and real value creation.

Guarantee reimbursements

Let’s say you want to borrow a cryptocurrency X, for example USDT, which will be considered as a representation of the dollar on the blockchain. This loan can be used to purchase real goods, particularly in the medium term where cryptocurrencies have developed and are more widely used as a means of exchange. A decentralized lending protocol will allow you, under conditions, to make this loan.

The schematic operation of such a protocol reflects that of a traditional bank. On the one hand, agents with excess liquidity will deposit their currency X in a pool, with the prospect of receiving interest. It’s the equivalent of going to deposit money in the bank. On the other hand, people needing to borrow will draw currency X from this pool, and will of course have to repay it later with interest. It’s the equivalent of getting a loan from the bank.

The supply of credit, that is to say the quantity of X made available by depositors, is generally not equal to the demand. For example, there may be 100 units of X deposited, and only 40 used by borrowers. In this case, we say that the utilization rate is 40%. The first reason why this rate is important is that it determines how much a depositor ultimately earns: if the borrowing rate is, for example, 5%, but only 40% of the liquidity is used, a depositor will obtain a return by 40% by 5%, or 2%. The second reason is that smart contracts typically specify the borrowing rate as a function of the usage rate. Choosing this feature is the first key feature of a lending protocol, since it determines the cost of borrowing for people who need it. How to make this choice optimally is a current active question in academic research, initiated by our work.

The second key feature of a lending protocol is what is called its collateralization rule. In traditional finance, a loan is guaranteed either by a real asset, the house it will be used to buy for example, or by the prospect of future income audited by the bank. In decentralized finance, agents are difficult to identify: how can we ensure that borrowers will actually have to repay?

A solution currently being provided is that of overcollateralization. To explain it, let’s go back to our example where we wanted to borrow X over a year, say for a value of 100 dollars. Let’s imagine that the interest rate is 5%, so that we will have to repay the equivalent of 105 dollars in X at maturity. The loan protocol will, for example, ask us to post as collateral, that is to say to block on the platform, at least the equivalent of 150 dollars in another cryptocurrency Y. Since the value of the collateral falls below 150, a liquidator will be able to intervene, sell the collateral for (approximately) 150, which will be sufficient to repay the 105 owed to the lender and also remunerate the liquidator.

With this procedure, lenders no longer face the risk of default, because borrowers provide collateral of sufficient value to repay the amount they owe. In practice, of course, a borrower will not want to make losses by being immediately liquidated, and will therefore put Y as collateral for a value of 300, for example. In this case liquidation will only take place in the event that Y loses half of its value.

Speculation objective?

This mechanism has the advantage of practically eliminating the risk of default, and therefore allowing lenders to calmly deposit their currency. But it has a major drawback: it does not really allow you to gain liquidity and offer credit with real added value. Indeed, to obtain a liquid asset, one must already hold another asset that is itself liquid, namely the collateral to be deposited.

What’s the point then? How to understand that Aave and Compound, the two main decentralized lending protocols today, govern liquidity pools with an aggregate value of several billion dollars (over 11 billion and 2 billion, respectively, as of February 27 2024)?

Loan protocols with overcollateralization offer above all the possibility of speculation. Suppose I think that currency Y will appreciate against X in the near future. So I would like to increase the amount of Y that I hold. To do this, I can deposit my Y as collateral to borrow X, then use these X to buy more Y. Since I think Y will appreciate in relation to to acquire the always, and I ultimately recover it to the extent that I repay my debt in

This strategy is nothing new: it is leverage, similar to that which one would carry out on a traditional stock market. The innovative aspect is that the decentralized lending protocol allows the implementation of this strategy in the context of cryptocurrencies.

Solution tokens?

But speculation remains a zero-sum game. Decentralized loans should be able to fuel real value creation, that is to say useful borrowing for those who need it without it being to the detriment of others. A key path to achieving this goal is tokenization.

The term tokenization comes from English token (token), which designates a unit of value on the blockchain: thus Ether is the “base” token on Ethereum. “Tokenizing” an asset means giving it representation on the blockchain. This allows you to benefit from the advantages of blockchain, such as transparency and immutability of data and transactions. Once tokenized, an asset can be used in a decentralized finance protocol.

In the context of lending, let’s say I’m a student with a car and I need liquidity while I wait to start a part-time job next month. If I can tokenize the title of this car, then this token can be used to overcollateralize a currency loan (which I intend to repay so as not to lose my car) and allow me to get through the coming month without any problems. The operation of the loan is as described above, but the fundamental difference is that this time we used illiquid collateral in order to obtain liquidity.

Tokenization already occupies a prominent place, whether in entrepreneurial practice or in the thinking of regulators and central banks. The American asset management giant, BlackRock, launched its first tokenized fund in March, BUIDL (BlackRock USD Institutional Digital Liquidity Fund), which represents on the blockchain a portfolio of dollars, American Treasury bills and other assets liquids. The Canadian company Polymath offers to tokenize real estate assets. Backed Finance, a young company based in Switzerland, is interested in traditional financial assets, such as Treasury bills or stock portfolios. In Argentina, Agrotoken converts agricultural commodities, such as wheat and soy, into tokens. And this is only a tiny part of the projects undertaken in this area.

However, the solution is not ideal. Indeed, it involves reintroducing centralization and intermediation: an entity must verify the existence and conformity of tokenized assets. We also lose anonymity because the token issuer must be identified. In doing so, we move away from the initial ideal of the cryptocurrency sphere, which was to build an entirely decentralized ecosystem, without intrusion from third-party authorities, and anonymous or pseudonymous.

However, this seems to be the price to pay to give a real role to decentralized lending, and hope to result in reliable and accessible borrowing. Cryptocurrencies would then make it possible to carry out at least a certain number of real asset transactions, and decentralized lending would have the potential to offer lower costs than traditional intermediaries and/or easier access to borrowing.

This article is part of the file “Decentralized finance: towards a world without intermediaries? », published by the online scientific media of the University of Paris Dauphine – PSL.