Systemic Risk in DeFi: A Network-Based Fragility Analysis of TVL Dynamics

TL;DR

Problem: Systemic risk in DeFi emerges from network interdependencies rather than isolated shocks. Approach: build time-varying correlation networks across protocol categories from TVL data; define Correlation Fragility Indicator (CFI) via PCA on four metrics $\bar{s}_t$, $\lambda_{\max,t}$, $d^{\text{strong}}_t$, and $H_t$, and decompose risk with the node-level Risk Contribution Score (RCS) defined as $\mathrm{RCS}_{i,t} = \mathrm{CFI}_t - \mathrm{CFI}_t^{(-i)}$. Findings: CFI tracks periods of rising aggregate liquidity instability and synchrony; top categories by RCS are not always the largest by TVL; targeted removal by RCS yields larger CFI reductions than random, especially in high-fragility regimes. Significance: the framework enables continuous ecosystem-level risk monitoring and stress testing for DeFi macroprudential design.

Abstract

Systemic risk refers to the overall vulnerability arising from the high degree of interconnectedness and interdependence within the financial system. In the rapidly developing decentralized finance (DeFi) ecosystem, numerous studies have analyzed systemic risk through specific channels such as liquidity pressures, leverage mechanisms, smart contract risks, and historical risk events. However, these studies are mostly event-driven or focused on isolated risk channels, paying limited attention to the structural dimension of systemic risk. Overall, this study provides a unified quantitative framework for ecosystem-level analysis and continuous monitoring of systemic risk in DeFi. From a network-based perspective, this paper proposes the DeFi Correlation Fragility Indicator (CFI), constructed from time-varying correlation networks at the protocol category level. The CFI captures ecosystem-wide structural fragility associated with correlation concentration and increasing synchronicity. Furthermore, we define a Risk Contribution Score (RCS) to quantify the marginal contribution of different protocol types to overall systemic risk. By combining the CFI and RCS, the framework enables both the tracking of time-varying systemic risk and identification of structurally important functional modules in risk accumulation and amplification.

Figures (7)

• Figure 1: DeFi correlation networks (snapshot at rolling-window end date: 11 May 2022). The left panel shows the fully weighted correlation network constructed from category-level TVL log returns over the rolling window, where each edge weight equals the absolute correlation $|C_{ij}|$. The right panel shows the thresholded network retaining only edges with $|C_{ij}| > 0.3$ to highlight the core of strong dependencies. Node size is proportional to category-level TVL, node color denotes node strength (sum of absolute correlations), edge color indicates signed correlation, and edge width reflects correlation magnitude.
• Figure 2: Time series of the standardized DeFi CFI based on rolling correlation networks of category-level TVL log returns. The series is oriented so that higher values indicate stronger network-wide synchronization and higher structural fragility.
• Figure 3: Systemically important protocol categories based on RCS. Panel (a) ranks protocol types by their average marginal contribution to system-wide fragility. Panel (b) reports the frequency with which each category appears in the top-10 RCS ranking across rolling windows.
• Figure 4: Attack curves: average CFI drop after removing $k$ nodes under targeted (RCS-based), strength-based, and random removal (random: 95% interval across dates).
• Figure 5: Attack curves in high-fragility vs. low-fragility regimes (defined by top/bottom CFI quantiles). Targeted removal is markedly more effective in high-fragility states.