Using Weighted Shapley Values to Measure the Systemic Risk of Interconnected Banks

Abstract

A macro‐prudential approach to financial regulation should reasonably evaluate the systemic risk of individual institutions. We introduce a measure of the systemic risk of interconnected banks using the weighted Shapley value, which characterizes banks' systemic risk by capturing two characteristics: (i) a bank's risk exposure of the primitive asset structure; and (ii) its position in the interbank network. An empirical implementation reveals that the primitive asset structure and interbank connections are main drivers of each bank's systemic risk. In addition, different types of interbank linkages (debt or share) make significant differences in affecting an individual bank's systemic risk.

Interbank money market concerns and actors’ strategies—A systematic review of 21st century literature

“…(2015), the Kenyan IMM by Fan et al. (2018), the Chinese banking system by Lin (2018), the European payment system TARGET2 by Gabrieli and Salakhova (2019), and the Austrian IMM network by Diem et al. (2020) all confirmed that systemic risk positively depends on the level of interconnectedness in the interbank network.…”

“…There is an idea that an interconnected bank’s systemic risk depends not only on the expected loss it imposes directly on depositors but also on its position in the IMM network (Lin, 2018; Puhr et al., 2012). Accordingly, the first defaulted node’s central position in the network is the next frequent attribute that affects the systemic risk.…”

“…Simulations performed by Calimani et al ( 2019) also suggested similar results. From the empirical point of view, results from the studies of the German IMM by Anand et al (2015), the Kenyan IMM by Fan et al (2018), the Chinese banking system by Lin (2018), the European payment system TARGET2 by Gabrieli and Salakhova (2019), and the Austrian IMM network by Diem et al (2020) all confirmed that systemic risk positively depends on the level of interconnectedness in the interbank network. One possible reason for these findings could be that with more connectivity, there are more paths through which contagion can spread.…”

“…Positive BB: Prefer to borrow from well-connected banks, if less-connected themselves, and vice versa LB: Prefer to lend to well-connected banks, if less-connected themselves, and vice versa (Anand et al, 2015;Caccioli et al, 2012;Calimani et al, 2019;Diem et al, 2020;Fan et al, 2018;Gabrieli & Salakhova, 2019;Gaffeo et al, 2019;Lenzu & Tedeschi, 2012;Lin, 2018;Simaan et al, 2020;Walters et al, 2019;Yao et al, 2016) Negative Similar to the positive effect (Duffy et al, 2019;Gaffeo et al, 2019;Li et al, 2019;Nguyen, 2018;Rünstler, 2016;Simaan et al, 2020;Walters et al, 2019) Network: centrality Positive BB and LB: Become more central as they are larger; more tighten with important counterparts CB: Considers core banks more important in a systemic sense (Degryse & Nguyen, 2007;Ferrara et al, 2019;Ka...…”

Valuing Inputs Under Supply Uncertainty: The Bayesian Shapley Value

“…The next example shows how the a priori Shapley value can be applied to a financial system of banks (see, e.g., Allen and Gale (2000), Battiston et al(2012), Lin (2016)) in order to measure the contribution of each bank to systemic risk. The risk measure function h is given by:…”