Financial risk contagion based on dynamic multi-layer network between banks and firms

How the ownership structures cause epidemics in financial markets: A network-based simulation model

This paper contributes to the finance literature by investigating the risk contagion and bailout effects in the global financial system. We use the bilateral exposure matrix to examine the impact of credit shocks and liquidity shocks on the global financial system when a Global Systemically Important Bank (G-SIB) is supposed to be in distress. The findings indicate that the global financial network is dominated by large and medium-sized financial institutions. Credit shocks does not play a significant role in risk contagion nor do they trigger the systemic risk in the global financial system. However, the global financial network is more vulnerable to liquidity shocks than to credit shocks. If a credit shock and liquidity shock overlaps with each other, the hybrid shock can exacerbate the risk contagion effects and lead to rapid propagation throughout the global financial network. If a regulator bails out the distressed institutions, the bailout efficiency increases with the number of rescued institutions. However, the improvement in of bailout efficiency decreases with the number of rescued institutions. If a regulator can promote financial institutions to raise the rollover ratio, the financial system will be more robust to shocks. This paper provides a ground view of systemic risks contagion and sheds light on directions for future research on systemic risk contagion in the global financial system. • The credit shock does not play a significant role in risk contagion in the global financial system. However, the liquidity shock has shown more substantial effects than that of the credit shock. • If the credit shock and liquidity shock overlaps with each other, the contagion effects are amplified. • If the regulator implements the bailout, the total rescue effects in the second round is better than that of other rounds. • If the financial institutions can raise the rollover ratio, the impact of systemic risk could be reduced significantly.

Financial systemic risk is an important issue in economics and financial systems. Trying to detect and respond to systemic risk with growing amounts of data produced in financial markets and systems, a lot of researchers have increasingly employed machine learning methods. Machine learning methods study the mechanisms of outbreak and contagion of systemic risk in the financial network and improve the current regulation of the financial market and industry. In this paper, we survey existing researches and methodologies on assessment and measurement of financial systemic risk combined with machine learning technologies, including big data analysis, network analysis and sentiment analysis, etc. In addition, we identify future challenges, and suggest further research topics. The main purpose of this paper is to introduce current researches on financial systemic risk with machine learning methods and to propose directions for future work.

Financial markets create endogenous systemic risk, the risk that a substantial fraction of the system ceases to function and collapses. Systemic risk can propagate through different mechanisms and channels of contagion. One important form of financial contagion arises from indirect interconnections between financial institutions mediated by financial markets. This indirect interconnection occurs when financial institutions invest in common assets and is referred to as overlapping portfolios. In this work we quantify systemic risk from indirect interconnections between financial institutions. Complete information of security holdings of major Mexican financial intermediaries and the ability to uniquely identify securities in their portfolios, allows us to represent the Mexican financial system as a bipartite network of securities and financial institutions. This makes it possible to quantify systemic risk arising from overlapping portfolios. We show that focusing only on direct interbank exposures underestimates total systemic risk levels by up to 50% under the assumptions of the model. By representing the financial system as a multi-layer network of direct interbank exposures (default contagion) and indirect external exposures (overlapping portfolios) we estimate the mutual influence of different channels of contagion. The method presented here is the first quantification of systemic risk on national scales that includes overlapping portfolios.

The Formation of Financial Networks

Determinants of risk in the banking sector during the European Financial Crisis

Systemic Risk Contribution from Financial Network in the UK

The multi-layer network nature of systemic risk and its implications for the costs of financial crises

Financial networks, bank efficiency and risk-taking

Granularity Instead of Entity-Centricity in the Financial Regulatory Data Architecture

Macro-Prudential Regulation from the Perspective of the Financial Network

Evolving interdependencies across institutions and markets drive systemic financial risk, yet most forecasting models either treat assets independently or rely on static correlation structures. This limitation becomes particularly salient as cryptocurrency markets increasingly interact with traditional banking systems amid financial stress. Ignoring time-varying cross-market network structure risks understating tail risk ‎precisely during periods when accurate systemic risk assessment is most critical. This study proposes a dynamic graph neural network ‎‎(GNN) framework for systemic risk forecasting that models time-varying financial networks spanning banking institutions and major cryptocurrency assets. Nodes represent financial entities, while edges are constructed using rolling-window dependency measures that adapt to ‎changing market conditions. Node dynamics are modeled through temporal neural architectures, and stress regimes are explicitly identified ‎to evaluate performance under market turmoil. The empirical design includes strong temporal baselines, static-graph ablations, and cross-market removal experiments to isolate the contribution of network dynamics and crypto-market integration. Results indicate that a strong ‎LSTM baseline achieves superior volatility forecasting accuracy in both overall and stress-period evaluations, demonstrating the competitiveness of purely temporal models. However, within the class of graph-based models, dynamic GNNs substantially outperform static-graph variants during stress periods, demonstrating the importance of time-varying network structure for capturing volatility amplification. ‎Bank-only and full-system dynamic GNNs exhibit comparable stress-period performance, suggesting that cryptocurrency assets contribute ‎limited incremental information to bank-specific forecasts, while remaining informative for system-level stress characterization. The findings ‎suggest that dynamic graph representations enhance stress sensitivity and structural interpretability relative to static network models, even ‎when they do not surpass strong temporal baselines in raw predictive accuracy. The results support a restrained view of crypto–banking ‎contagion, emphasizing its conditional relevance during periods of market stress rather than unconditional systemic dominance.

Spectral signatures of structural change in financial networks

We address a fundamental problem that is systematically encountered when modeling real-world complex systems of societal relevance: the limitedness of the information available. In the case of economic and financial networks, privacy issues severely limit the information that can be accessed and, as a consequence, the possibility of correctly estimating the resilience of these systems to events such as financial shocks, crises and cascade failures. Here we present an innovative method to reconstruct the structure of such partially-accessible systems, based on the knowledge of intrinsic node-specific properties and of the number of connections of only a limited subset of nodes. This information is used to calibrate an inference procedure based on fundamental concepts derived from statistical physics, which allows to generate ensembles of directed weighted networks intended to represent the real system—so that the real network properties can be estimated as their average values within the ensemble. We test the method both on synthetic and empirical networks, focusing on the properties that are commonly used to measure systemic risk. Indeed, the method shows a remarkable robustness with respect to the limitedness of the information available, thus representing a valuable tool for gaining insights on privacy-protected economic and financial systems.

The role of loan portfolio losses and bank capital for Asian financial system resilience