Currently, lending and borrowing in DeFi are facilitated through money market creators such as Aave and Compound. These big DeFi protocols were the first generation money markets that provided a peer-to-protocol model in which users can interact through a shared asset pool. In this model, users can provide collateral and then borrow any other asset that the protocol provides on their platform. However, this shared pool lending model comes with a significant risk as an exploit on one asset can cause a risk to the entire pool. Additionally, in this model, collateral is constantly being rehypothecated. Although this may create more liquidity in the markets, it can also cause destablization and lead to a riskier lending and borrowing environment which may deter users from using these services. This may be more capital efficient in certain cases, but I believe that lending and borrowing in DeFi is still scary to most users due to exploits and hacks.
Due to this risks of a shared pool lending model, new protocols developed that used different models to minimize risk such as the isolated asset pool model. These protocols allowed users to create their own pools with their own choice of asset pairings to isolate risk if a pool was to be exploited. This would mean that the exploit would not put the entire protocol at risk but only that certain pool. However, this does have a downside of fragmented liquidity as liquidity is spread between all the pools as they can be created by any user. Various protocols using the isolated pool model have been created that seem to be achieve less riskier lending and borrowing such as Compound III and Silo Finance. In the case with Compound III, users can borrow a single asset using different tokens as collateral and earn interest if they supply the borrowable asset. In the case with Silo Finance, users can deposit a single asset and use a bridge asset (ETH) to borrow other assets. These protocol designs may reduce risk for the user, but it also comes with some trade-offs. For example, with Silo Finance, because there is a bridge asset, users will essentially need to pay higher gas fees (several transactions) to borrow the asset they want. Nonetheless, the restructuring of lending and borrowing protocols to reduce risk for users and simplify complexities will allow for more users to onboard onto these DeFi services which will grow the user base.
Lending and borrowing in DeFi will come with trade-offs. There is no perfect solution that is risk-free and capital efficient with maximal liquidity. Future protocols will continue to experiment with different models to benefit users depending on what they want to do. I believe that this is good for the development and innovation of DeFi lending and borrowing protocols because these services need to be battle-tested in the real world to compete with traditional financial lending and borrowing services. As of right now, interacting with these DeFi protocols are scary for normal people. A lot of protocols seem too complex and only seem like they are for advanced traders that know the substantial risk they are taking on. However, to truly legitimitize the DeFi space to compete against traditional financial services, I believe that protocols need to cut down on complexities to make it easier for new users to onboard to use their services. Building a DeFi product that can provide multi-asset lending and borrowing initially can lead to various unseen risks and complexities that deter normal people away from DeFi. I believe that developers need to build starting from the basics especially since blockchain technology is relatively new. Building a simple DeFi lending and borrowing service with minimal risk can lead to a better tested product that others can then follow and innovate on which will lead to more complex products while attracting a bigger user base as growth occurs. However, this all is easier said than done in reality, but hopefully, DeFi developers keep reinventing ways of structuring financial services to truly change how global financial systems operate.
