Default cascades in complex networks: topology and systemic risk

Systemic risk measured by the resiliency of the system to initial shocks

The analysis of systemic risk often revolves around examining various measures utilized by practitioners and policymakers. These measures typically focus on assessing the extent to which external events can impact a financial system, without delving into the nature of the initial shock. In contrast, our approach takes a symmetrical standpoint and introduces a set of measures centered on the quantity of external shock that the system can absorb before experiencing deterioration. To achieve this, we employ a linearized version of DebtRank, which facilitates a clear depiction of the onset of financial distress, thereby enabling accurate estimation of systemic risk. Through the utilization of spectral graph theory, we explicitly compute localized and uniform exogenous shocks, elucidating their behavior. Additionally, we expand the analysis to encompass heterogeneous shocks, necessitating computation via Monte Carlo simulations. We firmly believe that our approach is both comprehensive and intuitive, enabling a standardized assessment of failure risk in financial systems.

Spatial Spillovers of Financial Risk and Their Dynamic Evolution: Evidence from Listed Financial Institutions in China

This paper investigates the multidimensional spatial effects of risk spillovers among Chinese financial institutions and the dynamic evolution of financial risk contagion in the tail risk correlation network over different time periods. We first measure risk spillovers from financial submarkets to the stock market, identifying five periods using structural breakpoint tests. Then, we construct a spatial error financial network panel model by combining complex network and spatial econometric theory to explore the spatial spillover variability. Finally, we calculate the Bonacich centrality of nodes in the tail risk network and analyze the dynamic evolution of the financial impact path during the different time periods. The results show that the multidimensional spatial spillovers of financial risk among financial institutions are obvious and time varying. The spatial spillovers of financial institutions are positively correlated with the turnover rate and negatively correlated with the exchange rate, interest rate and return volatility. Financial institutions of the same type in the tail risk network display intraindustry risk clustering, and the systemically important institutions identified based on Bonacich centrality differ significantly across time. Moreover, when risk spillovers increase, external shocks' destructive power and speed of transmission to the network rise.

Viewing the global health system as a complex adaptive system - implications for research and practice

The global health system (GHS) is ill-equipped to deal with the increasing number of transnational challenges. The GHS needs reform to enhance global resilience to future risks to health. In this article we argue that the starting point for any reform must be conceptualizing and studying the GHS as a complex adaptive system (CAS) with a large and escalating number of interconnected global health actors that learn and adapt their behaviours in response to each other and changes in their environment. The GHS can be viewed as a multi-scalar, nested health system comprising all national health systems together with the global health architecture, in which behaviours are influenced by cross-scale interactions. However, current methods cannot adequately capture the dynamism or complexity of the GHS or quantify the effects of challenges or potential reform options. We provide an overview of a selection of systems thinking and complexity science methods available to researchers and highlight the numerous policy insights their application could yield.   We also discuss the challenges for researchers of applying these methods and for policy makers of digesting and acting upon them. We encourage application of a CAS approach to GHS research and policy making to help bolster resilience to future risks that transcend national boundaries and system scales.

Network control by a constrained external agent as a continuous optimization problem

Social science studies dealing with control in networks typically resort to heuristics or solely describing the control distribution. Optimal policies, however, require interventions that optimize control over a socioeconomic network subject to real-world constraints. We integrate optimisation tools from deep-learning with network science into a framework that is able to optimize such interventions in real-world networks. We demonstrate the framework in the context of corporate control, where it allows to characterize the vulnerability of strategically important corporate networks to sensitive takeovers, an important contemporaneous policy challenge. The framework produces insights that are relevant for governing real-world socioeconomic networks, and opens up new research avenues for improving our understanding and control of such complex systems.

A model for cascading failures with the probability of failure described as a logistic function

In most cascading failure models in networks, overloaded nodes are assumed to fail and are removed from the network. However, this is not always the case due to network mitigation measures. Considering the effects of these mitigating measures, we propose a new cascading failure model that describes the probability that an overloaded node fails as a logistic function. By performing numerical simulations of cascading failures on Barabási and Albert (BA) scale-free networks and a real airport network, we compare the results of our model and the established model describing the probability of failure as a linear function. The simulation results show that the difference in the robustness of the two models depends on the initial load distribution and the redistribution of load. We further investigate the conditions of our new model under which the network exhibits the strongest robustness in terms of the load distribution and the network topology. We find the optimal value for the parameter of the load distribution and demonstrate that the robustness of the network improves as the average degree increases. The results regarding the optimal load distribution are verified by theoretical analysis. This work can be used to develop effective mitigation measures and design networks that are robust to cascading failure phenomena.

Controlling systemic risk: Network structures that minimize it and node properties to calculate it

Evaluation of systemic risk in networks of financial institutions in general requires information of interinstitution financial exposures. In the framework of the DebtRank algorithm, we introduce an approximate method of systemic risk evaluation which requires only node properties, such as total assets and liabilities, as inputs. We demonstrate that this approximation captures a large portion of systemic risk measured by DebtRank. Furthermore, using Monte Carlo simulations, we investigate network structures that can amplify systemic risk. Indeed, while no topology in general sense is a priori more stable if the market is liquid (i.e., the price of transaction creation is small) [T. Roukny et al., Sci. Rep. 3, 2759 (2013)10.1038/srep02759], a larger complexity is detrimental for the overall stability [M. Bardoscia et al., Nat. Commun. 8, 14416 (2017)10.1038/ncomms14416]. Here we find that the measure of scalar assortativity correlates well with level of systemic risk. In particular, network structures with high systemic risk are scalar assortative, meaning that risky banks are mostly exposed to other risky banks. Network structures with low systemic risk are scalar disassortative, with interactions of risky banks with stable banks.

Modeling financial distress propagation on customer-supplier networks

Financial networks have been the object of intense quantitative analysis during the last few decades. Their structure and the dynamical processes on top of them are of utmost importance to understand the emergent collective behavior behind economic and financial crises. In this paper, we propose a stylized model to understand the "domino effect" of distress in client-supplier networks. We provide a theoretical analysis of the model, and we apply it to several synthetic networks and a real customer-supplier network, supplied by one of the largest banks in Europe. Besides, the proposed model allows us to investigate possible scenarios for the functioning of the financial distress propagation and to assess the economic health of the full network. The main novelty of this model is the combination of two stochastic terms: an additive noise, accounting by the capability of trading and paying obligations, and a multiplicative noise representing the variations of the market. Both parameters are crucial to determining the maximum default probability and the diffusion process characteristics.

The hyperbolic geometry of financial networks

Based on data from the European banking stress tests of 2014, 2016 and the transparency exercise of 2018 we construct networks of European banks and demonstrate that the latent geometry of these financial networks can be well-represented by geometry of negative curvature, i.e., by hyperbolic geometry. Using two different hyperbolic embedding methods, hydra+ and Mercator, this allows us to connect the network structure to the popularity-vs-similarity model of Papdopoulos et al., which is based on the Poincaré disc model of hyperbolic geometry. We show that the latent dimensions of 'popularity' and 'similarity' in this model are strongly associated to systemic importance and to geographic subdivisions of the banking system, independent of the embedding method that is used. In a longitudinal analysis over the time span from 2014 to 2018 we find that the systemic importance of individual banks has remained rather stable, while the peripheral community structure exhibits more (but still moderate) variability. Based on our analysis we argue that embeddings into hyperbolic geometry can be used to monitor structural change in financial networks and are able to distinguish between changes in systemic relevance and other (peripheral) structural changes.

Loan maturity aggregation in interbank lending networks obscures mesoscale structure and economic functions

Since the 2007-2009 financial crisis, substantial academic effort has been dedicated to improving our understanding of interbank lending networks (ILNs). Because of data limitations or by choice, the literature largely lacks multiple loan maturities. We employ a complete interbank loan contract dataset to investigate whether maturity details are informative of the network structure. Applying the layered stochastic block model of Peixoto (2015) and other tools from network science on a time series of bilateral loans with multiple maturity layers in the Russian ILN, we find that collapsing all such layers consistently obscures mesoscale structure. The optimal maturity granularity lies between completely collapsing and completely separating the maturity layers and depends on the development phase of the interbank market, with a more developed market requiring more layers for optimal description. Closer inspection of the inferred maturity bins associated with the optimal maturity granularity reveals specific economic functions, from liquidity intermediation to financing. Collapsing a network with multiple underlying maturity layers or extracting one such layer, common in economic research, is therefore not only an incomplete representation of the ILN's mesoscale structure, but also conceals existing economic functions. This holds important insights and opportunities for theoretical and empirical studies on interbank market functioning, contagion, stability, and on the desirable level of regulatory data disclosure.

Consensus and polarization in competing complex contagion processes

The rate of adoption of new information depends on reinforcement from multiple sources in a way that often cannot be described by simple contagion processes. In such cases, contagion is said to be complex. Complex contagion happens in the diffusion of human behaviours, innovations and knowledge. Based on that evidence, we propose a model that considers multiple, potentially asymmetric and competing contagion processes and analyse its respective population-wide dynamics, bringing together ideas from complex contagion, opinion dynamics, evolutionary game theory and language competition by shifting the focus from individuals to the properties of the diffusing processes. We show that our model spans a dynamical space in which the population exhibits patterns of consensus, dominance, and, importantly, different types of polarization, a more diverse dynamical environment that contrasts with single simple contagion processes. We show how these patterns emerge and how different population structures modify them through a natural development of spatial correlations: structured interactions increase the range of the dominance regime by reducing that of dynamic polarization, tight modular structures can generate structural polarization, depending on the interplay between fundamental properties of the processes and the modularity of the interaction network.

Flickering in Information Spreading Precedes Critical Transitions in Financial Markets

As many complex dynamical systems, financial markets exhibit sudden changes or tipping points that can turn into systemic risk. This paper aims at building and validating a new class of early warning signals of critical transitions. We base our analysis on information spreading patterns in dynamic temporal networks, where nodes are connected by short-term causality. Before a tipping point occurs, we observe flickering in information spreading, as measured by clustering coefficients. Nodes rapidly switch between "being in" and "being out" the information diffusion process. Concurrently, stock markets start to desynchronize. To capture these features, we build two early warning indicators based on the number of regime switches, and on the time between two switches. We divide our data into two sub-samples. Over the first one, using receiver operating curve, we show that we are able to detect a tipping point about one year before it occurs. For instance, our empirical model perfectly predicts the Global Financial Crisis. Over the second sub-sample, used as a robustness check, our two statistical metrics also capture, to a large extent, the 2016 financial turmoil. Our results suggest that our indicators have informational content about a future tipping point, and have therefore strong policy implications.

Bond percolation in coloured and multiplex networks

Percolation in complex networks is a process that mimics network degradation and a tool that reveals peculiarities of the network structure. During the course of percolation, the emergent properties of networks undergo non-trivial transformations, which include a phase transition in the connectivity, and in some special cases, multiple phase transitions. Such global transformations are caused by only subtle changes in the degree distribution, which locally describe the network. Here we establish a generic analytic theory that describes how structure and sizes of all connected components in the network are affected by simple and colour-dependent bond percolations. This theory predicts locations of the phase transitions, existence of wide critical regimes that do not vanish in the thermodynamic limit, and a phenomenon of colour switching in small components. These results may be used to design percolation-like processes, optimise network response to percolation, and detect subtle signals preceding network collapse.

Percolation is a tool used to investigate a network’s response as random links are removed. Here the author presents a generic analytic theory to describe how percolation properties are affected in coloured networks, where the colour can represent a network feature such as multiplexity or the belonging to a community.

A multiplex financial network approach to policy evaluation: the case of euro area Quantitative Easing

Over the last decades, both advanced and emerging economies have experienced a striking increase in the intra-financial activity across different asset classes and increasingly complex contract types, leading to a far more complex financial system. Until the 2007-2008 crisis, the increased financial intensity and complexity was believed beneficial in making the financial system more resilient and less vulnerable to shocks. However, in 2007-2008, the advanced economies suffered the biggest financial crisis since the 1930s, followed by a severe post-crisis recession, questioning the adequacy of traditional tools in predicting, explaining, and responding to periods of financial distress. In particular, the effect of complex interconnections among financial actors on financial stability has been widely acknowledged. A recent debate focused on the effects of unconventional policies aimed at achieving both price and financial stability. Among these unconventional policies, Quantitative Easing (QE, i.e., the large-scale asset purchase programme conducted by a central bank upon the creation of new money) has been recently implemented by the European Central Bank (ECB). In this context, two questions deserve more attention in the literature. First, to what extent, the resources provided to the banking system through QE are transmitted to the real economy. Second, to what extent, the QE may also alter the pattern of intra-financial exposures and what are the implications in terms of financial stability. Here, we address these two questions by developing a methodology to map the multilayer macro-network of financial exposures among institutional sectors across financial instruments (i.e., loans and deposits, debt securities, and equity), and we illustrate our approach on recently available data. We then test the effect of the implementation of ECB's QE on the time evolution of the financial linkages in the multilayer macro-network of the euro area, as well as the effect on macroeconomic variables, such as consumption, investment, unemployment, growth, and inflation.

Correlations between thresholds and degrees: An analytic approach to model attacks and failure cascades

Two node variables determine the evolution of cascades in random networks: a node's degree and threshold. Correlations between both fundamentally change the robustness of a network, yet they are disregarded in standard analytic methods as local tree or heterogeneous mean field approximations, since order statistics are difficult to capture analytically because of their combinatorial nature. We show how they become tractable in the thermodynamic limit of infinite network size. This enables the analytic description of node attacks that are characterized by threshold allocations based on node degree. Using two examples, we discuss possible implications of irregular phase transitions and different speeds of cascade evolution for the control of cascades.

Modelling indirect interactions during failure spreading in a project activity network

Spreading broadly refers to the notion of an entity propagating throughout a networked system via its interacting components. Evidence of its ubiquity and severity can be seen in a range of phenomena, from disease epidemics to financial systemic risk. In order to understand the dynamics of these critical phenomena, computational models map the probability of propagation as a function of direct exposure, typically in the form of pairwise interactions between components. By doing so, the important role of indirect interactions remains unexplored. In response, we develop a simple model that accounts for the effect of both direct and subsequent exposure, which we deploy in the novel context of failure propagation within a real-world engineering project. We show that subsequent exposure has a significant effect in key aspects, including the: (a) final spreading event size, (b) propagation rate, and (c) spreading event structure. In addition, we demonstrate the existence of 'hidden influentials' in large-scale spreading events, and evaluate the role of direct and subsequent exposure in their emergence. Given the evidence of the importance of subsequent exposure, our findings offer new insight on particular aspects that need to be included when modelling network dynamics in general, and spreading processes specifically.

Reverse stress testing interbank networks

We reverse engineer dynamics of financial contagion to find the scenario of smallest exogenous shock that, should it occur, would lead to a given final systemic loss. This reverse stress test can be used to identify the potential triggers of systemic events, and it removes the arbitrariness in the selection of shock scenarios in stress testing. We consider in particular the case of distress propagation in an interbank market, and we study a network of 44 European banks, which we reconstruct using data collected from banks statements. By looking at the distribution across banks of the size of smallest exogenous shocks we rank banks in terms of their systemic importance, and we show the effectiveness of a policy with capital requirements based on this ranking. We also study the properties of smallest exogenous shocks as a function of the parameters that determine the endogenous amplification of shocks. We find that the size of smallest exogenous shocks reduces and that the distribution across banks becomes more localized as the system becomes more unstable.

Distress Propagation in Complex Networks: The Case of Non-Linear DebtRank

We consider a dynamical model of distress propagation on complex networks, which we apply to the study of financial contagion in networks of banks connected to each other by direct exposures. The model that we consider is an extension of the DebtRank algorithm, recently introduced in the literature. The mechanics of distress propagation is very simple: When a bank suffers a loss, distress propagates to its creditors, who in turn suffer losses, and so on. The original DebtRank assumes that losses are propagated linearly between connected banks. Here we relax this assumption and introduce a one-parameter family of non-linear propagation functions. As a case study, we apply this algorithm to a data-set of 183 European banks, and we study how the stability of the system depends on the non-linearity parameter under different stress-test scenarios. We find that the system is characterized by a transition between a regime where small shocks can be amplified and a regime where shocks do not propagate, and that the overall stability of the system increases between 2008 and 2013.

Leveraging the network: A stress-test framework based on DebtRank

We develop a novel stress-test framework to monitor systemic risk in financial systems. The modular structure of the framework allows to accommodate for a variety of shock scenarios, methods to estimate interbank exposures and mechanisms of distress propagation. The main features are as follows. First, the framework allows to estimate and disentangle not only first-round effects (i.e. shock on external assets) and second-round effects (i.e. distress induced in the interbank network), but also third-round effects induced by possible fire sales. Second, it allows to monitor at the same time the impact of shocks on individual or groups of financial institutions as well as their vulnerability to shocks on counterparties or certain asset classes. Third, it includes estimates for loss distributions, thus combining network effects with familiar risk measures such as VaR and CVaR. Fourth, in order to perform robustness analyses and cope with incomplete data, the framework features a module for the generation of sets of networks of interbank exposures that are coherent with the total lending and borrowing of each bank. As an illustration, we carry out a stress-test exercise on a dataset of listed European banks over the years 2008–2013. We find that second-round and third-round effects dominate first-round effects, therefore suggesting that most current stress-test frameworks might lead to a severe underestimation of systemic risk.

How damage diversification can reduce systemic risk

We study the influence of risk diversification on cascading failures in weighted complex networks, where weighted directed links represent exposures between nodes. These weights result from different diversification strategies and their adjustment allows us to reduce systemic risk significantly by topological means. As an example, we contrast a classical exposure diversification (ED) approach with a damage diversification (DD) variant. The latter reduces the loss that the failure of high degree nodes generally inflict to their network neighbors and thus hampers the cascade amplification. To quantify the final cascade size and obtain our results, we develop a branching process approximation taking into account that inflicted losses cannot only depend on properties of the exposed, but also of the failing node. This analytic extension is a natural consequence of the paradigm shift from individual to system safety. To deepen our understanding of the cascade process, we complement this systemic perspective by a mesoscopic one: an analysis of the failure risk of nodes dependent on their degree. Additionally, we ask for the role of these failures in the cascade amplification.

Enhancing speed of pinning synchronizability: low-degree nodes with high feedback gains

Controlling complex networks is of paramount importance in science and engineering. Despite recent efforts to improve controllability and synchronous strength, little attention has been paid to the speed of pinning synchronizability (rate of convergence in pinning control) and the corresponding pinning node selection. To address this issue, we propose a hypothesis to restrict the control cost, then build a linear matrix inequality related to the speed of pinning controllability. By solving the inequality, we obtain both the speed of pinning controllability and optimal control strength (feedback gains in pinning control) for all nodes. Interestingly, some low-degree nodes are able to achieve large feedback gains, which suggests that they have high influence on controlling system. In addition, when choosing nodes with high feedback gains as pinning nodes, the controlling speed of real systems is remarkably enhanced compared to that of traditional large-degree and large-betweenness selections. Thus, the proposed approach provides a novel way to investigate the speed of pinning controllability and can evoke other effective heuristic pinning node selections for large-scale systems.

Mapping Systemic Risk: Critical Degree and Failures Distribution in Financial Networks

The financial crisis illustrated the need for a functional understanding of systemic risk in strongly interconnected financial structures. Dynamic processes on complex networks being intrinsically difficult to model analytically, most recent studies of this problem have relied on numerical simulations. Here we report analytical results in a network model of interbank lending based on directly relevant financial parameters, such as interest rates and leverage ratios. We obtain a closed-form formula for the “critical degree” (the number of creditors per bank below which an individual shock can propagate throughout the network), and relate failures distributions to network topologies, in particular scalefree ones. Our criterion for the onset of contagion turns out to be isomorphic to the condition for cooperation to evolve on graphs and social networks, as recently formulated in evolutionary game theory. This remarkable connection supports recent calls for a methodological rapprochement between finance and ecology.

DebtRank: A Microscopic Foundation for Shock Propagation

The DebtRank algorithm has been increasingly investigated as a method to estimate the impact of shocks in financial networks, as it overcomes the limitations of the traditional default-cascade approaches. Here we formulate a dynamical "microscopic" theory of instability for financial networks by iterating balance sheet identities of individual banks and by assuming a simple rule for the transfer of shocks from borrowers to lenders. By doing so, we generalise the DebtRank formulation, both providing an interpretation of the effective dynamics in terms of basic accounting principles and preventing the underestimation of losses on certain network topologies. Depending on the structure of the interbank leverage matrix the dynamics is either stable, in which case the asymptotic state can be computed analytically, or unstable, meaning that at least one bank will default. We apply this framework to a dataset of the top listed European banks in the period 2008-2013. We find that network effects can generate an amplification of exogenous shocks of a factor ranging between three (in normal periods) and six (during the crisis) when we stress the system with a 0.5% shock on external (i.e. non-interbank) assets for all banks.

Failure dynamics of the global risk network

Risks threatening modern societies form an intricately interconnected network that often underlies crisis situations. Yet, little is known about how risk materializations in distinct domains influence each other. Here we present an approach in which expert assessments of likelihoods and influence of risks underlie a quantitative model of the global risk network dynamics. The modeled risks range from environmental to economic and technological, and include difficult to quantify risks, such as geo-political and social. Using the maximum likelihood estimation, we find the optimal model parameters and demonstrate that the model including network effects significantly outperforms the others, uncovering full value of the expert collected data. We analyze the model dynamics and study its resilience and stability. Our findings include such risk properties as contagion potential, persistence, roles in cascades of failures and the identity of risks most detrimental to system stability. The model provides quantitative means for measuring the adverse effects of risk interdependencies and the materialization of risks in the network.

Using trading strategies to detect phase transitions in financial markets

We show that the log-periodic power law singularity model (LPPLS), a mathematical embodiment of positive feedbacks between agents and of their hierarchical dynamical organization, has a significant predictive power in financial markets. We find that LPPLS-based strategies significantly outperform the randomized ones and that they are robust with respect to a large selection of assets and time periods. The dynamics of prices thus markedly deviate from randomness in certain pockets of predictability that can be associated with bubble market regimes. Our hybrid approach, marrying finance with the trading strategies, and critical phenomena with LPPLS, demonstrates that targeting information related to phase transitions enables the forecast of financial bubbles and crashes punctuating the dynamics of prices.

Credit Default Swaps networks and systemic risk

Credit Default Swaps (CDS) spreads should reflect default risk of the underlying corporate debt. Actually, it has been recognized that CDS spread time series did not anticipate but only followed the increasing risk of default before the financial crisis. In principle, the network of correlations among CDS spread time series could at least display some form of structural change to be used as an early warning of systemic risk. Here we study a set of 176 CDS time series of financial institutions from 2002 to 2011. Networks are constructed in various ways, some of which display structural change at the onset of the credit crisis of 2008, but never before. By taking these networks as a proxy of interdependencies among financial institutions, we run stress-test based on Group DebtRank. Systemic risk before 2008 increases only when incorporating a macroeconomic indicator reflecting the potential losses of financial assets associated with house prices in the US. This approach indicates a promising way to detect systemic instabilities.

Cascades on a stochastic pulse-coupled network

While much recent research has focused on understanding isolated cascades of networks, less attention has been given to dynamical processes on networks exhibiting repeated cascades of opposing influence. An example of this is the dynamic behaviour of financial markets where cascades of buying and selling can occur, even over short timescales. To model these phenomena, a stochastic pulse-coupled oscillator network with upper and lower thresholds is described and analysed. Numerical confirmation of asynchronous and synchronous regimes of the system is presented, along with analytical identification of the fixed point state vector of the asynchronous mean field system. A lower bound for the finite system mean field critical value of network coupling probability is found that separates the asynchronous and synchronous regimes. For the low-dimensional mean field system, a closed-form equation is found for cascade size, in terms of the network coupling probability. Finally, a description of how this model can be applied to interacting agents in a financial market is provided.

Percolation on fitness-dependent networks with heterogeneous resilience

The ability to understand the impact of adversarial processes on networks is crucial to various disciplines. The objects of study in this article are fitness-driven networks. Fitness-dependent networks are fully described by a probability distribution of fitness and an attachment kernel. Every node in the network is endowed with a fitness value and the attachment kernel translates the fitness of two nodes into the probability that these two nodes share an edge. This concept is also known as mutual attractiveness. In the present article, fitness does not only serve as a measure of attractiveness, but also as a measure of a node's robustness against failure. The probability that a node fails increases with the number of failures in its direct neighborhood and decreases with higher fitness. Both static and dynamic network models are considered. Analytical results for the percolation threshold and the occupied fraction are derived. One of the results is that the distinction between the dynamic and the static model has a profound impact on the way failures spread over the network. Additionally, we find that the introduction of mutual attractiveness stabilizes the network compared to a pure random attachment.