Financial stress relationships among Euro area countries: an R-vine copula approach

ABSTRACT

One of the biggest challenges of keeping Euro area financial stability is the negative co-movement between the vulnerability of public finance, the financial sector, security markets stresses as well as economic growth, especially in peripheral economies. This paper utilizes a ARMA-GARCH based R-vine copula method to explore tail dependance between the Financial Stress Indices of 11 euro area countries with an aim of understanding how financial stress are interacting with each other. We find larger economies in the Euro area tend to have closer upper tail dependence in terms of positive shocks, while smaller economies tend to have closer lower tail dependence with respect to negative shocks. The R-vine copula results underline the complex dynamics of financial stress relations existing between Euro Area economies. The estimated R-vine shows Spain, Italy, France and Belgium are the most inter-connected nodes which underlying they might be more efficient targets to treat in order to achieve a quicker stabilizing. Our results relate to the fact that Eurozone is not a unified policy making area, therefore, it needs to follow divergent policies for taming the effects of financial instability to different regions or groups of economies that are more interconnected.

KEYWORDS:

• Financial stress
• FSI
• Euro area
• GARCH
• vine copula
• regular vine

JEL CLASSIFICATIONS:

• E44
• F47
• G01

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Given the significant decline in the GDP of majority of European economies, triggered by the global financial crisis 2008–2009, the annual GDP growth rate in European Union declined from to from 2008 to 2009.

2 They use 1 for stress period and 0 for non-stress period to define the financial conditions in an economy.

3 Kliesen and Smith (2010) construct 11 financial market variables from US (Kansas City Financial Stress Index) to measure the degree of financial stress in the markets. Hanschel and Monnin (2005) forecast Switzerland banking sector crisis by using stress index. Morales and Estrada (2010) use FSI for determining bank profitability and probability of default in Colombia. Yiu et al. (2010) use a monthly composite FSI to capture historical episodes when Hong Kong was under significant financial stress.

4 Systemic risk, in turn, can be defined as the risk that financial instability becomes so widespread that it impairs the functioning of a financial system to the point where economic growth and welfare suffer materially.

5 Table 4 shows the unit root test results from Augmented Dickey–Fuller (ADF) test (Dickey and Fuller 1979), Phillips–Perron (PP) test (Phillips and Perron 1988) and DF-GLS test (Elliott, Rothenberg, and Stock 1996) of each country's FSI. FSIs from Finland, Greece and Netherlands passes PP tests while fails ADF tests and DF-GLS tests, and all other series fails all three tests which suggests they are non-stationary with at least one unit root. Both tests results from first difference of FSIs show no more unit root. Therefore, by balancing ADF, PP and DF-GLS test results, first difference is applied to all series.

6 where k is the number of estimated parameters, L is likelihood, n is sample size in order to find the best fitted model (Schwarz 1978).

7 From January 2009 to November 2014, there are clear diverse-movement of 10-year sovereign bond yields between German and Ireland, Portugal, Spain and Greece during peak time of European Sovereign Debt Crisis 2010–2013 (ECB 2014).

8 The log-likelihood value for C-vine is 1017.193 from Equation (18(18) ). The log-likelihood value for D-vine is 1056.477 from Equation (21(21) ). And the log-likelihood value for R-vine is 1070.483 from Equation (23(23) ).