Contents lists available at ScienceDirect Economic Modelling journal homepage: www.elsevier.com/locate/econmod Contagion risk for Australian banks from global systemically important banks: Evidence from extreme events ☆ Selim Akhter, Kevin Daly ⁎ ,1 Discipline of Economics, Finance and Property School of Business, Western Sydney University, Australia ARTICLE INFO JEL Classification: G21 G210 Keywords: Global systemically important banks (GSIBs) Australian banks Extreme value theory (EVT) Extreme events Distance to default (DD) GARCH Logistic regression model ABSTRACT This paper presents evidence that extreme negative shocks for the global systemically important banks (GSIBs) are contagious to Australian banks. Our logit regression models predict transmission of adverse extreme shocks in the distance to default (DD) of GSIBs to the Australian banks. While most previous studies consider contagion across national stock markets, we investigate the degree of contagion risk for Australian banks spreading from GSIBs. Our results point to the critical importance for the Australian Prudential Regulation Authority (APRA) (2015) to closely observe and monitor developments across the major GSIBs and direct appropriate local policy measures accordingly. 1. Introduction and background This paper investigates the degree of contagion risk facing Australian banks spreading from global systemically important US, European and Japanese banks. We define contagion risk for Australian banks as the transmission of extreme negative shocks from a group of global systemically important banks (GSIBs). Our definition of con- tagion is similar to that used by governments, citizens, and policy- makers as the fear that negative events in another country, outside of their regulatory controls, can spread and have deleterious effects for the home country. We identify extreme negative shocks by changes in the distance to default (DD), where DD measures the distance between the present value of a bank's assets and their liabilities (described in more details in the following section). Hence, a larger DD for a bank is indicative of a stronger financial position while a smaller DD indicates financial distress or weakness. We estimate DD for eight Australian owned banks and twenty GSIBs on a daily basis and compute the daily change in DD (ΔDD) over a seven-year timeframe. We then isolate all significant negative shocks from the time series for each bank's ΔDD before proceeding to estimate the probability of these negative events spreading to Australian banks. Historically, Australia's banks have maintained a relatively healthy and stable financial position compared to their overseas counterparts. Whilst the 2007-08 global financial crises had a marked negative impact on GSIBs and Australia's banks, Australia's financial markets generally performed better than other developed countries markets. For example, a major trigger of the global financial crisis (GFC) was the widespread availability and use of sub-prime mortgages along with failures to correctly assess counterparty risk, both of which were controlled and monitored to a greater degree by Australia's regulatory authorities compared to other developed countries (Guy 2009). Australian banks survived the GFC with no announced bank failures and only a slight increase in nonperforming loans (IMF 2012). The profitability and capital adequacy ratios (CAR) of Australia's authorized depository institutions (ADIs) experienced a steady increase over 2008-15. Fig. 1 shows the CAR of ADIs rose from 11.4 percent in December 2007 to 13.1 percent in the quarter ending June 2015. Return on Equity (ROE) also increased after a decline during the GFC, with annualized after-tax return on equity recorded at 18.04 percent in June 2015 increasing from 4.5 percent in September 2009. The strengths of the Australian financial environment do not however exclude future threats as rightly perceived by the Financial System Inquiry (FSI) (2014). Here the FSI observed that the Australian financial system has characteristics giving rise to particular risks, in particular its dependence on imported capital (FSI Final Report 2014: http://dx.doi.org/10.1016/j.econmod.2016.11.018 Received 17 February 2016; Received in revised form 22 November 2016; Accepted 24 November 2016 ☆ We are also grateful to the anonymous reviewers for their valuable and constructive comments on an earlier version of this paper. ⁎ Corresponding author. 1 This author wishes to dedicate the paper to my co-author Dr Selim Akhter who suddenly passed away in September 2017, my sincere condolences to Selim’s Family and Colleagues. E-mail addresses: s.akhter@westernsydney.edu.au (S. Akhter), k.daly@westernsydney.edu.au (K. Daly). Economic Modelling 63 (2017) 191–205 0264-9993/ © 2017 Elsevier B.V. All rights reserved. MARK