[Paper Review] SoK: Decentralized Finance (DeFi) -- Fundamentals, Taxonomy and Risks

Krzysztof Gogol, Christian Killer+4 more|arXiv (Cornell University)|Apr 17, 2024

Banking stability, regulation, efficiencyEconomics, Econometrics and Finance3 citations

TL;DR

This paper proposes a systematic taxonomy and risk analysis framework for Decentralized Finance (DeFi) protocols, classifying them into three core categories—liquidity pools, synthetic/pegged tokens, and aggregators—based on design and architecture. It identifies that DeFi risk stems from protocol design, token type, and usage, with synthetic assets like liquid staking tokens facing significant de-peg risk, especially under market stress.

ABSTRACT

Decentralized Finance (DeFi) refers to financial services that are not necessarily related to crypto-currencies. By employing blockchain for security and integrity, DeFi creates new possibilities that attract retail and institution users, including central banks. Given its novel applications and sophisticated designs, the distinction between DeFi services and understanding the risk involved is often complex. This work systematically presents the major categories of DeFi protocols that cover over 90\% of total value locked (TVL) in DeFi. It establishes a structured methodology to differentiate between DeFi protocols based on their design and architecture. Every DeFi protocol is classified into one of three groups: liquidity pools, pegged and synthetic tokens, and aggregator protocols, followed by risk analysis. In particular, we classify stablecoins, liquid staking tokens, and bridged (wrapped) assets as pegged tokens resembling similar risks. The full risk exposure of DeFi users is derived not only from the DeFi protocol design but also from how it is used and with which tokens.

Motivation & Objective

• To establish a structured, evidence-based taxonomy for DeFi protocols based on their architectural and functional design.
• To analyze the risk exposure of DeFi users by examining protocol design, token types, and usage patterns.
• To identify and classify the major DeFi categories that account for over 85% of Total Value Locked (TVL).
• To highlight systemic risks, particularly de-peg risk in synthetic and pegged tokens, including liquid staking and stablecoins.
• To demonstrate that risk in DeFi is multidimensional, involving agents, protocols, and the specific tokens involved.

Proposed method

• Classify DeFi protocols into three main categories: liquidity pool protocols, synthetic/pegged token protocols, and aggregator protocols based on functional and architectural characteristics.
• Use data from DeFi Llama to identify and analyze the top DeFi categories by Total Value Locked (TVL), focusing on those with over $5B TVL.
• Analyze risk dimensions through a three-layered framework: risk from DeFi agents (e.g., lenders, LPs), protocol design, and token-specific risks (e.g., de-peg, slashing).
• Examine synthetic and pegged tokens—including stablecoins, wrapped assets, and liquid staking tokens—focusing on their de-peg vulnerabilities and design trade-offs.
• Assess the role of on-chain settlement and blockchain security in shaping DeFi risk, particularly in relation to MEV and staking incentives.
• Evaluate the impact of protocol design on network security, including the risk of staking withdrawal due to yield arbitrage in lending vs. staking.

Figure 1: Value Proposition and Design of DeFi Protocols

Experimental results

Research questions

• RQ1 How can DeFi protocols be systematically classified based on their core design and function, and which categories dominate the current ecosystem in terms of TVL?
• RQ2 What are the primary risk dimensions affecting DeFi users, and how do these risks vary across different types of protocols and tokens?
• RQ3 Why are synthetic and pegged tokens—such as liquid staking and stablecoins—particularly vulnerable to de-pegging, and what factors influence this risk?
• RQ4 To what extent do aggregator protocols amplify or redistribute risk by combining exposure to multiple underlying protocols and tokens?
• RQ5 How does the design of DeFi protocols, especially in relation to staking and yield generation, impact blockchain network security and validator behavior?

Key findings

• The three dominant DeFi protocol categories—liquidity pools, synthetic/pegged tokens, and aggregators—account for over 85% of total DeFi TVL as of August 2023.
• Liquid staking protocol Lido surpassed MakerDAO in TVL in 2023, becoming the largest DeFi protocol, highlighting the growing dominance of synthetic and pegged assets.
• Synthetic and pegged tokens, including liquid staking and stablecoins, are at significant risk of de-pegging due to volatility, design flaws, and the inclusion of staking rewards in their value proposition.
• Liquidity providers (LPs) and service users face distinct risk profiles, with LPs exposed to impermanent loss and protocol-specific risks, while users face counterparty and execution risks.
• Aggregator protocols inherit and concentrate the risks of the underlying protocols and tokens they interact with, making them systemic risk amplifiers.
• Over-collateralized lending protocols expose borrowers to liquidation risk, and yield-seeking behavior can undermine blockchain security by incentivizing stakers to exit staking for higher yields in DeFi protocols.

Figure 2: Taxonomy of blockchain-based financial application