DeFi 2.0
This article is the third part of a three-part series on “Decentralized Finance – a Hype, a Threat or an Opportunity for Regulated Financial Institutions?” .
Since DeFi Summer 2020, DeFi applications have become a significant trend in the blockchain industry. Total Value Locked (TVL), the value of digital assets blocked in DeFi applications such as Ethereum, has increased in just two years from around 1 bn. USD in mid-2020 to around 69 bn. USD on August 05, 2022, although a collapse in the price of many tokens has caused the TVL to move away significantly from the peak of 254 Mia. USD in December 2021[1]. Decentralized financial infrastructure has been shown to enable interesting new business models, such as decentralized exchanges (DEX) or automated asset management (AAM). So far, investors can achieve high returns on their invested capital within these business models, but they also take corresponding risks, which especially inexperienced investors often do not recognize. The combination of returns and supposedly low risk has attracted a great many investors, primarily early adopters from the retail segment.
The first part of this series of articles presented how decentralized finance (DeFi) and centralized finance (CeFi) differ, how trust can be created through smart contracts even without intermediaries, and how various business models such as decentralized exchanges and automated asset managers function. The second part deals with the opportunities and challenges that decentralized finance opens up for banks, such as the opportunity to expand their own range of digital assets with protocol or governance tokens from DeFi applications or to act as a facilitator for investors who do not have sufficient background knowledge to invest in DeFi business models on their own. Both articles highlight strengths and potentials of DeFi applications, e.g., the independence from central intermediaries, which prevents the system from having a central point of attack and strengthens the investors’ power of disposal over their assets, or yield farming[2], which enables investors to increase their investment without additional capital investment compared to traditional investment opportunities.
However, vulnerabilities of DeFi applications have also been identified over the past 24 months. For example, the borrowing/lending platform Compound 2021 discovered a bug in the smart contract that allowed investors to request governance tokens from Compound (“COMP”) without providing liquidity to the lending pool. The bug was closed by the development team and nearly half of the stolen $114 million was returned to Compound[3]. In addition to technical errors such as incorrectly programmed smart contracts, human error, e.g. in the form of poor coverage of one’s stablecoin with digital assets or very risky investment strategies with borrowed capital, has also led to the demise of some DeFi applications and CeFi companies since the first post was published in April 2022, e.g. Terraform Labs with the stablecoin Terra Luna[4] and CeFi companies, such as the borrowing/lending service provider Celsius[5] or the crypto hedge fund 3 Arrows Capital[6].
The third and final part takes a closer look at recent developments in the DeFi sector and how they address well-known challenges in the context of DeFi, such as the potential for errors when setting up smart contracts, the lack of incentive structures for investors, or the requirements for investors’ technical and professional knowledge, without compromising the strengths of DeFi applications.
Challenges of DeFi 1.0[7]
Lack of Incentive Structures
The revenue opportunities in DeFi business models are diverse, e.g., liquidity providers on decentralized exchanges (DEX) receive a share of trading fees for the liquidity they provide. Likewise, lenders receive a fee for providing capital in decentralized borrowing/lending applications. The search for the maximum possible return in the context of DeFi applications is referred to as yield farming. In addition to the returns generated by, for example, providing digital assets, investors often receive tokens of the DeFi application, which they can trade on exchanges. Investors often choose the DeFi application with the current maximum return from yield farming and liquidity mining and quickly transfer digital assets to another application if it can offer better conditions, e.g., by paying higher rewards for providing capital. In DeFi 1.0, there are almost no incentives for investors to commit to a DeFi application for a longer period of time, which in turn drives up the liquidity costs for protocol operators as higher incentives for liquidity providers have to be provided by the DeFi protocol.
Scalability
Another challenge is the scalability of DeFi applications. The scalability of a blockchain is part of the so-called blockchain trilemma, which describes the area of tension between scalability, decentralization and security. By giving a higher weighting to one factor, the other two factors (in this case, for example, security and decentralization) are negatively affected[8]. Scalability describes how many transactions an application can process with predominantly constant speed and transaction costs (gas fees). It depends primarily on the scalability of the underlying layer-1 blockchain. Layer-1 blockchain refers to blockchain protocols on which further applications are built. These applications use the underlying architecture of the layer-1 blockchain (e.g. Ethereum) with all its advantages and disadvantages to run their own application (e.g. DEX). One of the main reasons DeFi applications are difficult to scale is the use of the Ethereum blockchain as the settlement layer for transactions. In this context, the term settlement refers to the posting of transactions on the blockchain. This includes querying the holdings of the sending public key, mining the block with the transaction details, attributing them to the receiving public key and communicating (“broadcasting”) the completed, correct transaction to the other blockchain participants and nodes. The immutability of the recorded transactions on the settlement layer is essential for the automatic processing of smart contracts since transactions are not checked again by a (central) intermediary before execution. The Ethereum settlement layer is thus the single source of truth for all transactions of DeFi applications that use the Ethereum blockchain as a basis. A majority of DeFi 1.0 applications are built on Ethereum due to the availability of smart contracts as well as the stable technical framework.
The Ethereum blockchain, as the second largest blockchain by market capitalization[9], has a very strong decentralization; however, the high transaction costs as well as the comparatively slow block times, for example, stand in the way of scaling DeFi applications[10]. The transaction costs are determined by the available storage space per block as well as the block time (the duration of time until a new block is attached to a blockchain). The smaller the available storage space per block as well as the higher the block time, the greater the cost per transaction. The Ethereum blockchain wants to take a first step towards scalability with the upcoming change of the consensus mechanism from proof-of-work to proof-of-stake[11].
Understanding DeFi
Handling one’s own private key, transactions to and from DeFi applications, and participation in DeFi business models is complex and requires prior knowledge. Additional specialized technical and financial knowledge is often necessary for successful participation in DeFi business models. For example, what are the implications of changing price bands when trading pairs are submitted to DEX, or how are governance tokens most effectively used for yield farming? The former affects revenue opportunities for liqudity providers, as price bands set the financial framework for their own offering (buy & sell price of the provided trading pair). Governance tokens represent a key revenue component of DeFi business models and can be further lent as part of liquidity mining. Without a fundamental understanding of DeFi as well as the underlying DeFi business models, the risk for investors increases significantly.
Decentralization
The decentralization of DeFi applications depends on the one hand on how many tokens are held by individual parties, e.g., the developers or larger investors, and how many are actually in free circulation. If a large proportion of tokens is held by a small number of blockchain participants, these participants can, for example, strongly influence the further development of the protocol or influence the price when the tokens held are sold. Second, decentralization depends on who has control over the nodes that participate in the consensus mechanism. In particular, new DeFi applications with their own blockchain run the initial nodes all by themselves until a large enough community has formed. This de facto centralization of nodes with the development team raises serious questions about the degree of decentralization of new DeFi applications with their own blockchain.
In order for DeFi applications to experience broader adoption and become suitable for mass use, the weaknesses or optimization potentials of DeFi 1.0 presented here must be eliminated. DeFi 2.0 attempts to do just that with the help of some new approaches.
DeFi 2.0
DeFi 2.0 aims to further expand the strengths from DeFi 1.0 and address its existing weaknesses. This will involve both technical solutions to address the weaknesses (e.g. blockchain-based smart contract insurance) and new incentive systems for investors to keep liquidity in liquidity or lending pools for longer. This section presents possible solutions and, where possible, provides a practical example.
Smart Contracts
Programmed smart contracts are used to process decentralized financial transactions without central intermediaries. If the smart contract contains errors or is attacked by third parties, the digital assets blocked in it are at risk. Decentralized smart contract insurance has been developed to ensure the recovery of at least some of the assets in the event of an attack or a faulty smart contract. The insurance works similarly to a classic insurance in that the necessary funds are made up of the insured’s contributions and certain conditions must be met for injured parties to recover a portion of the assets lost. An example of such smart contract insurance is Nexus Mutual. Unlike a traditional insurance policy, which strictly adheres to the terms and conditions set out in the contract, Nexus Mutual’s Claims Assessment Committee members decide whether to grant the request for payment of the insured amount[12]. In principle, any token holder can act as a claims assessor. For this purpose, the claims assessor stakes a certain amount of the Nexus token for a predefined period of time. To ensure that claims assessors behave honorably, an advisory committee watches over their activities. In case of fraudulent behavior, claims assessors can lose their entire stock of staked Nexus tokens[13].
New Incentive Systems for Investors
As an incentive for investors to keep digital assets longer in a DeFi application, new incentive schemes are used in DeFi 2.0. One example is OlympusDAO, a DeFi project that aims to establish a stable currency in the DeFi sector. To achieve this, OlympusDAO assumes a role for its own currency OHM similar to that of a national bank in a classical economy: it controls the money supply, injects new money into the market or withdraws money from the market to stabilize the exchange rate. OHM tokens are backed 100% by the stablecoin DAI. OlympusDAO buys DAI tokens from investors, instead of investors lending the digital assets to the pool (as in a classic DeFi 1.0 application). These assets are used for the treasury of the OHM tokens. Investors can withdraw the OlympusDAO token OHM for an agreed discount after a certain time period (“bonding”)[14]. Delayed OHM payout reduces the likelihood that investors will exchange the OHM token directly for another asset upon receipt, thus draining liquidity from the protocol. Unlike a stablecoin such as USDC or Tether, Olympus’ claim is not to maintain a 1:1 peg to a legal currency, but to ensure that its holdings are backed by a digital asset. This is to allow some value to be attached to the OHM token. In practice, the price of the OHM token carries a price premium that reflects the high staking rewards for staking OHM tokens and the underlying treasury in DAI[15].
Scaling
As addressed in the first part of this post, a major obstacle to scaling DeFi applications is the limited scalability of the Ethereum blockchain. First steps towards a solution are scheduled for the near future by transitioning the Ethereum blockchain from proof-of-work (PoW) to proof-of-stake (PoS). In addition, shard chains and the Beacon Chain will create a new structure for the Ethereum blockchain (expected to be available in 2023). Shard chains are additional chains that each process a portion of the transactions on a blockchain. By splitting transactions across shards, transactions become processable in parallel, helping to scale the Ethereum blockchain. Shard chains are referred to as an on-chain scaling solution and are part of the layer-1 solution[16].
In addition, layer-2 solutions enable scaling of the underlying layer-1 blockchain. Examples of layer 2 solutions include sidechains (see the Polygon example in this section) or zk rollups (zero knowledge rollups), which aggregate transactions into badges and process them collectively[17].
An example of the use of sidechains is the Polygon blockchain[18]. Sidechains are blockchains that have their own consensus mechanism and often their own tokens. Transactions made on the sidechain are first bundled and processed there and then transferred to the main chain. The Polygon blockchain secures its transactions on the Ethereum blockchain[19]. Within the Polygon blockchain, digital assets can be transferred at will and due to the more scalable consensus mechanism proof-of-stake, these transfers take place faster and cheaper than if the Ethereum blockchain were used for these transactions. In order to continue to benefit from the security advantages of the Ethereum blockchain (e.g. decentralization, consensus mechanism, number of nodes, etc.), the information about the transactions on the Polygon blockchain is also published on the Ethereum blockchain. By processing the actual transaction on the Polygon blockchain, the scaling obstacles on the Ethereum blockchain (as mentioned e.g., the high gas fees, the longer block times etc.) are basically solved. Well-known DeFi applications that use the Polygon blockchain include DEX Curve.fi or Aave’s decentralized borrowing-lending service, which I already discussed in the second post in this series[20].
User Experience
Interaction with DeFi applications requires a basic knowledge of blockchain technology. For example, users must know how transactions are carried out with their own private keys or how digital assets can be transferred across different blockchains. The complexity of DeFi business models, as well as the multiple blockchains that underlie them, makes managing one’s DeFi assets a time-consuming and complex activity. Increasing DeFi usage thus requires detailed technical knowledge among the general population or a simplification in dealing with DeFi applications and one’s own digital assets. The German company “Unstoppable Finance”[21] wants to achieve the latter by offering a wallet for end users that maps different blockchains and enables transfers between different blockchains. In this way, digital assets can be easily transferred between different blockchains and the current holdings across blockchains can be viewed in one place.
Decentralization
In addition to scaling, the degree of decentralization of a blockchain is another component of the blockchain trilemma discussed above[22]. The more decentralized a blockchain is, the higher the security, since a larger number of independent participants confirm the transactions on the blockchain. On the other hand, high decentralization slows down the transaction time due to more necessary confirmations. Another challenge of a high degree of decentralization is updates as well as further developments of the blockchain, which the individual (decentralized) token holders have to vote on before implementation. For example, the the introduction of SegWit (reduction of the block size and thus a lower block time) on the Bitcoin blockchain was already proposed in 2015, but the implementation with a soft fork only started in 2017. Since soft forks are compatible with older versions of the same blockchain, the original Bitcoin blockchain remains valid and participants can decide which standard to use for a transaction[23].
The tendentially lower number of transactions per second (TPS) of a highly decentralized blockchain can be mitigated by appropriate consensus mechanisms (e.g. proof-of-stake) or by a suitable blockchain architecture. Hence, also the conversion of the Ethereum blockchain. New layer-1 blockchains such as Cardano (“ADA”) or Solana (“SOL”) would like to position themselves as an alternative to Ethereum and enable scaling of DeFi applications through a higher number of TPS as well as the proof-of-stake consensus mechanism with comparatively equal security. This is achieved, among other things, by an appropriate architecture of the underlying blockchain, e.g., the Cardano blockchain consists of two parts: The settlement layer for processing transactions[24] and the computation layer, which provides the rules for processing transactions, e.g., rules for programmable tokens[25]. By dividing the activities into two layers, transactions can be executed more quickly on the settlement layer. Cardano in particular claims that all blocks on the Cardano blockchain are created by community staking pools, ensuring decentralized distribution of staking rewards[26]. If the opposite were true, i.e., if Cardano developers or early investors operated a portion of the staking pools, the protocol could not be said to be consistently decentralized. As it turns out, Cardano also has potential for improvement in the context of decentralization, as about 16 percent of all available ADA tokens – ADA is the native token of the Cardano blockchain – are under the control of a few investors[27]. Compared to the number of DeFi applications on the Ethereum blockchain, Cardano has a lot of catching up to do; currently there are only 73 DeFi applications on Cardano[28], compared to 203 on Ethereum[29].
Conclusion
In summary, it can be said that challenges of DeFi 1.0 are addressed by the DeFi industry through adapted mechanisms and aggregations with a high user experience. Nevertheless, these solutions are only a beginning, they are often limited to individual protocols and do not yet represent a market standard. The main challenge is ensuring decentralization, which has not yet been completely solved by DeFi 2.0. It will be exciting to see how the relationship between return and risk will settle in DeFi applications. It can be assumed that both returns and risks will decrease with the increasing maturity of the DeFi industry. The second article in this series touched on the topic of regulation, which it is imperative to address for greater DeFi adoption.
Current market movements and the bankruptcies of large DeFi applications such as Celsius and hedge fund 3 Arrow Capital are creating negative influences on prices and public perception of DeFi. However, in the medium to long term, these cases are very valuable for the DeFi industry as they are beneficial to the industry’s maturity through risk mitigation measures or smart contract-based investment safety nets.
Conclusion of the Blog Series on Decentralized Finance
Since DeFi Summer 2020, decentralized finance has outgrown its infancy and become a permanent fixture in the blockchain industry. Banks face a wide range of opportunities and challenges in dealing with the supposed disruption. Nevertheless, developments are still in the early stages and mass adoption of DeFi among retail and business customers has not yet been achieved. Regardless of the current market situation, the underlying concept of DeFi is relevant for banks and addressing the issue is essential. Exogenous factors such as regulation will continue to mature the DeFi industry, and addressing the impact on one’s own business model early on can protect against any unpleasant surprises.
[1] DeFi Llama, 2022 (DeFi Llama, 2022); abgerufen am 5.8.2022, von https://defillama.com/
[2] Zur Erinnerung: Yield Farming bezeichnet die Verleihung von Token zur Generierung von Zinseinkünften. Für das Zurverfügungstellen von Kapital erhalten die Investoren wiederum Token, die sie an anderer Stelle ebenfalls verleihen können.
[3] 13 Biggest DeFi Hacks and Heists, 2022 (Decrypt, 2022); Abgerufen am 2.7.2022, von https://decrypt.co/93874/biggest-defi-hacks-heists
[4] Weitere Informationen: Coindesk, 2022; Abgerufen am 1.7.2022, von https://www.coindesk.com/learn/the-fall-of-terra-a-timeline-of-the-meteoric-rise-and-crash-of-ust-and-luna/
[5] Weitere Informationen: Coindesk, 2022; Abgerufen am 1.7.2022, von https://www.coindesk.com/business/2022/06/30/crypto-lender-celsius-network-exploring-options-to-preserve-and-protect-assets/
[6] Weitere Informationen: Coindesk, 2022; Abgerufen am 1.7.2022, von https://www.coindesk.com/business/2022/06/29/genesis-faces-hundreds-of-millions-in-losses-as-3ac-exposure-swamps-crypto-lenders-sources/
[7] DeFi 1.0 und DeFi 2.0 sind keine trennscharfen Begriffe, sondern stehen für die erste Welle an DeFi-Anwendungen, deren Schwachstellen mittlerweile gut bekannt und dokumentiert sind (DeFi 1.0) und eine Reihe neuerer Anwendungen, die versuchen, eben diese Schwachstellen zu beheben.
[8] Das Blockchain Trilemma (Crypto Valley Journal, 2020) ; Abgerufen am 7.7.2022, von https://cvj.ch/fokus/hintergrund/das-blockchain-trilemma/
[9] Coinmarketcap, 2022 (Coinmarketcap, 2022); Abgerufen am 2.7.2022, von https://coinmarketcap.com/
[10] Why Ethereum Users Tolerate Exorbitant Gas Fees (coindesk.com, 2022) ; Abgerufen am 5.7.2022, von https://www.coindesk.com/layer2/2022/02/17/why-ethereum-users-tolerate-exorbitant-gas-fees/
[11] Die Zusammenführung (Ethereum.org); Abgerufen am 8.8.2022, von https://ethereum.org/de/upgrades/merge/
[12] Nexus Mutual (Nexus Mutual, 2022); Abgerufen am 2.7.2022, von https://nexusmutual.io/
[13] Claims Assessment (Nexus Mutual, 2021); Abgerufen am 15.5.2022, von https://nexusmutual.gitbook.io/docs/claims-assessment/claims-assessment
[14] Bonding (Olympus, 2022); Abgerufen am 2.7.2022, von https://docs.olympusdao.finance/main/basics/bonding
[15] Coinmarketcap (Coinmarketcap, 2022); Abgerufen am 10.8.2022, von https://coinmarketcap.com/currencies/olympus/
[16] Scaling (Ethereum.org, 2022); Abgerufen am 10.8.2022, von https://ethereum.org/en/developers/docs/scaling/
[17] Blockchain Layer 1 vs. Layer 2 Scaling Solutions (Binance, 2022); Abgerufen am 10.8.2022, von https://academy.binance.com/en/articles/blockchain-layer-1-vs-layer-2-scaling-solutions
[18] Polygon (Polygon, 2022); Abgerufen am 5.7.2022, von https://polygon.technology/solutions/polygon-pos/
[19] Sidechains (Ethereum.org, 2022); Abgerufen am 5.7.2022, von https://ethereum.org/en/developers/docs/scaling/sidechains/
[20] Top Polygon DeFi Apps (DappRadar, 2022); Abgerufen am 5.7.2022, von https://dappradar.com/rankings/protocol/polygon/category/defi
[21] Why DeFi needs a curation layer : introducing the Unstoppable Finance protocol valuation framework (Unstoppable Finance, 2022) ; Abgerufen am 5.7.2022, von https://medium.com/@unstoppablefinance/why-defi-needs-a-curation-layer-introducing-the-unstoppable-finance-protocol-evaluation-framework-b55fdae29603
[22] Das Blockchain Trilemma (Crypto Valley Journal, 2020); Abgerufen am 7.7.2022, von https://cvj.ch/fokus/hintergrund/das-blockchain-trilemma/
[23] Was ist SegWit (Seggregated Witness) und wie funktioniert es? (Bitpanda, 2022); Abgerufen am 10.8.2022, von https://www.bitpanda.com/academy/de/lektionen/was-ist-segwit-segregated-witness-und-wie-funktioniert-es/
[24] Designing In Layers – Cardano Settlement Layer (Why Cardano, 2020); Abgerufen am 10.8.2022, von https://why.cardano.org/en/introduction/designing-in-layers/
[25] Cardano Computation Layer (Why Cardano, 2020) ; Abgerufen am 10.8.2022, von https://why.cardano.org/en/introduction/cardano-computation-layer/
[26] Twitter (Input Output, 2021); Abgerufen am 5.7.2022, von https://twitter.com/InputOutputHK/status/1377376420540735489?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E1377376420540735489%7Ctwgr%5E%7Ctwcon%5Es1_&ref_url=https%3A%2F%2Ffinance.yahoo.com%2Fnews%2Fcardano-blockchain-achieves-100-decentralization-220615285.html
[27] Cardano-Wale halten über 16 Prozent aller ADA (BTC-Echo, 2022); Abgerufen am 5.7.2022, von https://www.btc-echo.de/schlagzeilen/cardano-wale-halten-ueber-16-prozent-aller-ada-138382/
[28] Dapps (CardanoCrowd, 2022); Abgerufen am 5.7.2022, von https://cardanocrowd.com/dapps
[29] Ethereum DeFi Ecosystem (defiprime.com, 2022); Abgerufen am 5.7.2022, von https://defiprime.com/ethereum#:~:text=We%20have%20226%20DeFi%20projects,of%20them%20built%20on%20Ethereum.
- Positioning opportunities for banks in the context of digital assets – A market observation - 13.02.2024
- Tokenization – Potentials, Challenges and Use Cases in the Environment of the Financial Industry – Part 2 - 06.06.2023
- Tokenization – Potentials, Challenges and Use Cases in the Financial Industry Environment - 14.02.2023