Summary:
Banks' common holding of mark-to-market assets can lead to their interconnectedness, and a negative shock-induced fire sale can give rise to risk contagion among such a network of banks, which is a main driver of systemic risk. We investigate how such interconnectedness drives systemic risk by using a theoretical network model and conducting an empirical analysis based on a sample of Chinese listed commercial banks. We find that the network is robust yet fragile. In our model, high interconnectedness among banks is equivalent to a high level of common risk exposure to the same asset class, and thus contributes more to systemic risk. We define this as the “risk contagion” effect of interconnectedness, which implies that the network is fragile. A high level of interconnectedness can also reduce banks' fire sales, as each bank holds each asset more lightly and thus will suffer fewer losses in a period of distress. Higher interconnectedness can therefore alleviate systemic risk by reducing the initial shocks received by each bank. We define this as the “risk sharing” effect of interconnectedness, which indicates that the network is robust. We also confirm that this risk sharing effect is only significant when a bank's mark-to-market asset portfolio is highly concentrated (a high HHI index), and when it faces large shocks. In contrast, if the portfolio is well diversified or the shock is small, the benefits of risk sharing will be weak, and then the risk contagion effect will dominate. Our empirical analysis provides further evidence in support of these theories. We collect data and our sample of commercial banks from WIND and the banks' financial report footnotes. For the baseline fixed-effects regression, we find a one standard deviation increase in interconnectedness leads to a 2.6% increase in systemic risk, indicating an overall risk contagion effect. In addition, when a bank's portfolio is highly concentrated and faces large shocks, the magnitude becomes-1.7%, indicating a risk sharing effect. We then use the amount of bond issuance as an instrumental variable for interconnectedness and regard the implementation of a new financial instrument (i.e., accounting rules) in China as a natural experiment, to confirm the causality of the relationship between interconnectedness and systemic risk. Both identification methodologies provide strong causal evidence. In addition, we find that interconnectedness reduces initial shocks when these shocks are large and banks' HHI indices are high, which is consistent with the prediction of the theoretical model. Finally, risk contagion is more pronounced for small and capital-insufficient banks, and risk sharing is more pronounced for large and more geographically diversified banks. Several meaningful implications for regulators can be drawn from our findings. First, any macroprudential regulation should take common holding interconnectedness into consideration. Our finding that sharing occurs under specific criteria can be helpful when facing a trade-off between the “robustness” and “fragility” of a financial network. Second, regulators can implement flexible policies such as open market operations to encourage banks to adjust their connectedness contingent on the level and concentration of shocks. Finally, to initially deter risk contagion, banks should improve their risk management and regulators can extend their security and currency market span while providing supplementary capital channels to strengthen the banks' anti-risk capabilities. This paper contributes to three strands of literature. First, we document the “robust yet fragile” feature of the indirect common holding network. Other studies examine either the robustness and fragility of direct interbank lending networks or the fragility of indirect networks, but the robustness of the latter is underexplored. We fill this research gap from both theoretical and empirical perspectives. We also identify the specific condition required for the risk sharing effect (robustness), which can inform the implementation of regulations. Second, we contribute to studies of asset fire sales in the financial market and common holdings by institutional investors. Numerous studies attribute fire sales to institutional investors putting pressure on asset prices and liquidity, thus generating systemic financial risk. However, few studies focus on how to tackle fire sales. We address this question by identifying the risk sharing effect of interconnectedness. Finally, our findings provide further evidence for how fair value accounting affects systemic risk, by examining the impact of a new accounting rule first implemented in China in 2018. This evidence contributes to the debate concerning the procyclicality of fair value accounting and its contribution to systemic risk.
Acemoglu, D., A. Ozdaglar and A. Tahbaz-Salehi, 2015, “Systemic Risk and Stability in Financial Networks”, The American Economic Review, 105(2), 564～608.
[13]
Adrian, T. and M. K. Brunnermeier, 2016, “CoVaR”, The American Economic Review, 106(7), 1705.
[14]
Allen, F. and D. Gale, 2000, “Financial Contagion”, Journal of Political Economy, 108(1), 1～33.
[15]
Benoit, S., J. E. Colliard, C. Hurlin and C. Pérignon, 2017, “Where the Risks Lie: A Survey on Systemic Risk”, Review of Finance, 21(1), 109～152.
[16]
Cabrales, A., P. Gottardi and F. Vega-Redondo, 2017, “Risk Sharing and Contagion in Networks”, The Review of Financial Studies, 30(9), 3086～3127.
[17]
Chu, Y., S. Deng and C. Xia, 2020, “Bank Geographic Diversification and Systemic Risk”, The Review of Financial Studies, 33(10), 4811～4838.
[18]
Cont, R. and E. Schaanning, 2019, “Monitoring Indirect Contagion”, Journal of Banking & Finance, 104, 85～102.
[19]
Duarte, F. and T. M. Eisenbach, 2021, “Fire-sale Spillovers and Systemic Risk”, The Journal of Finance, 76(3), 1251～1294.
[20]
Falato, A., A. Hortacsu, D. Li and C. Shin, 2021, “Fire-Sale Spillovers in Debt Markets”, The Journal of Finance, 76(6), 3055～3102.
[21]
Gatev, E. and P. E. Strahan, 2009, “Liquidity Risk and Syndicate Structure”, Journal of Financial Economics, 93(3), 490～504.
[22]
Girardi, G., K. W. Hanley, S. Nikolova, L. Pelizzon and M. G. Sherman, 2021, “Portfolio Similarity and Asset Liquidation in the Insurance Industry”, Journal of Financial Economics, 142(1), 69～96.
[23]
Greenwood, R., A. Landier and D. Thesmar, 2015, “Vulnerable Banks”, Journal of Financial Economics, 115(3), 471-485.
[24]
Plantin, G., H. Sapra and H. S. Shin, 2008, “Marking-to-market: Panacea or Pandora's Box？”, Journal of Accounting Research, 46(2), 435～460.
