Social dynamics of financial networks

Abstract

The global financial crisis in [2007][2008][2009] demonstrated that systemic risk can spread all over the world through a complex web of financial linkages, yet we still lack fundamental knowledge about the evolution of the financial web. In particular, interbank credit networks shape the core of the financial system, in which a time-varying interconnected risk emerges from a massive number of temporal transactions between banks. The current lack of understanding of the mechanics of interbank networks makes it difficult to evaluate and control systemic risk. Here, we uncover fundamental dynamics of interbank networks by seeking the patterns of daily transactions between individual banks. We find stable interaction patterns between banks from which distinctive network-scale dynamics emerge. In fact, the dynamical patterns discovered at the local and network scales share common characteristics with social communication patterns of humans. To explain the origin of "social" dynamics in interbank networks, we provide a simple model that allows us to generate a sequence of synthetic daily networks characterized by the observed dynamical properties. The discovery of dynamical principles at the daily resolution will enhance our ability to assess systemic risk and could contribute to the real-time management of financial stability.

Network models of financial systemic risk: a review

“…Lowering the time resolution (such as weeks or months) tends to increase the likelihood that a core-periphery structure is detected, yet a non-negligible fraction of such networks are still best characterized by a bipartite structure. Kobayashi and Takaguchi [21] reinforce their result by showing that the bipartivity of daily interbank networks have been increasing over the past decade. 11 If we regard interbank markets as dynamic systems in which the structure of bilateral exposures changes every day, an interesting question to ask is whether the daily dynamics are just random or there are robust and time-invariant properties.…”

“…11 If we regard interbank markets as dynamic systems in which the structure of bilateral exposures changes every day, an interesting question to ask is whether the daily dynamics are just random or there are robust and time-invariant properties. Kobayashi and Takaguchi [21] find that the daily market activity represented by combination (N, M), where N and M are the numbers of active banks and edges, respectively, is strictly ruled by a superlinear relationship N / M 1:5 , or hNi / ffiffiffiffi ffi M p , independently of the structure and the size of daily networks (Fig. 6a).…”

“…Since edges exposed to the risk of bank default are out-going edges (i.e., lending to other banks), the mechanism of social contagion applies here in a straightforward manner as long as there are no bidirectional edges. 6 The trick here is that the volumes of interbank exposures, or edge weights, are evenly distributed between 6 The presence of bidirectional edges, although they rarely appear in empirical financial networks [21], could affect the size of contagion [22,23].…”

“…In interbank markets, most of the bilateral transactions are overnight, meaning that the relationship between two banks as a lender and a borrower lasts only for one day or shorter, depending on the time when the loan contract is made. In the Italian interbank market, for example, more than 86% of transactions are overnight lending in the period between 2000 and 2015 [21]. If the issue of interest is to understand the interconnectedness of financial risks, then an aggregate network of overnight lending relationships contains irrelevant information because different edges formed in different days in fact do not exist at the same time.…”

Extracting the multi-timescale activity patterns of online financial markets

“…As social communications between humans form temporal social networks, financial transactions between banks also shape time-varying networks [9][10][11][12][13]. In the interbank market, for instance, overnight bilateral lending and borrowing between banks organize temporal networks whose structure changes on a daily basis, because the overnight financial contracts last for only one day [11,12]. Thus, financial markets, similarly to social networks, can be interpreted as complex systems where each agent's activity should be captured and characterized at appropriate time scales [11,12].…”