Global financial crisis and multiscale systematic risk: Evidence from selected European stock markets

• Author: Mohammad S. Hasan,Antonios K. Alexandridis
• Date: 01 October 2020
• DOI: http://doi.org/10.1002/ijfe.1764
• Published date: 01 October 2020

RESEARCH ARTICLE

Global financial crisis and multiscale systematic risk:

Evidence from selected European stock markets

Antonios K. Alexandridis | Mohammad S. Hasan

Kent Business School, University of Kent,

Canterbury, UK

Correspondence

Antonios K. Alexandridis, Kent Business

School, University of Kent, Canterbury,

Kent CT2 7PE, UK.

Email: a.alexandridis@kent.ac.uk

Abstract

In this paper, we have investigated the impact of the global financial crisis on

the multihorizon nature of systematic risk and market risk using daily data of

eight major European equity markets over the period of 2005–2018. The

method is based on a wavelet multiscale approach within the framework of a

capital asset pricing model. Empirical results demonstrate that beta coefficients

have a multiscale tendency, and betas tend to increase at higher scales (lower

frequencies). In addition, the size of betas and R

2

s tends to increase during the

crisis period compared with the precrisis period. The multiscale nature of the

betas is consistent with the fact that stock market investors have different time

horizons due to different trading strategies. Our results based on scale depen-

dent value at risk (VaR) suggest that market risk tends to be more concen-

trated at lower time scales (higher frequencies) of the data. Moreover, the

scale-by-scale estimates of VaR have increased almost three folds for every

market during the crisis period compared with the precrisis period. Finally,

our approach allows for accurately forecasting time-dependent betas and VaR.

KEYWORDS

CAPM, global financial crisis, multiscale systematic risk, wavelet analysis, wavelet networks

JEL CLASSIFICATION

C22; G15

1|INTRODUCTION

In this paper, we investigate the impact of the global

financial crisis on the multihorizon nature of systematic

risk and market risk. We use daily data from eight major

European equity markets. Our method is based on a

recent and powerful method to estimate both the market

risk and the systematic risk within the framework of the

CAPM using wavelet analysis (WA).

Although the major financial U.S. institutions, such

as New Century Financial, United States holding of

HSBC, and the world's top five investment banks suffered

huge losses in the subprime mortgage and collateralized

debt obligation transactions by Summer (2007), the world

financial system observed a period of relative calm with

some optimism regarding the outcome of the ongoing cri-

sis until the 8 months of 2008

1,2

. Figure 1 presents a cur-

sory example of several major banks' exposures to AIG

during the time of financial crisis for readers to under-

stand the magnitude and extent of the problem inherent

in the systemic risks associated with the financial system

and institutions.

The subprime mortgage crisis eventually erupted

when first, major U.S. financial firms, such as Lehman

Brothers and AIG, and then European financial institu-

tions, such as Northern Rock, Fortis, Dexia, and a

Received: 31 December 2017 Revised: 25 January 2019 Accepted: 13 September 2019

DOI: 10.1002/ijfe.1764

Int J Fin Econ. 2019;1–29. wileyonlinelibrary.com/journal/ijfe © 2019 John Wiley & Sons, Ltd. 1

518 © 2019 John Wiley & Sons, Ltd. Int J Fin Econ. 2020;25:518–546.wileyonlinelibrary.com/journal/ijfe

number of Icelandic banks, showed signs of insolvency.

3

The crisis exposed the inherent vulnerabilities, systemic

risks, and a catalogue of regulatory failures in the global

financial services industries. The meltdown of the sub-

prime crisis of 2007 exerted a meteor shower effect

across the world's stock market by the fourth quarter of

the 2008. In the last quarter of 2008, the stock markets

of both developed and emerging economies experienced

large decline in prices of securities.

4

Figure 2 presents

movements of stock market indices in the United States

and five European countries namely, Netherlands,

United Kingdom, Germany, Greece, and Spain, during

the period of 2005–2012, which uniformly demonstrates

a sharp decline of share prices during the period of

September–November, 2008 for all countries. Although

the stock markets of United Kingdom, Netherlands, and

Germany exhibit an upward trend after November 2008,

the stock markets of Greece and Spain show a persistent

downward trend in their share prices. It is clearly evi-

dent that the global financial crisis exerts an adverse

impact on both systematic and market risks for these

countries.

Eichengreen, Mody, Nedeljkovic, and Sarno (2012)

investigated the impact of subprime crisis on the global

banking system using a dynamic factor model. The study

employed principal components analysis to identify com-

mon factors in the movement of banks' credit default

swap spread (CDS). The study found that the share of the

variance accounted by common factors rose steadily to

exceptional level from the outbreak of the subprime cri-

sis, which reflected the heightened funding and counter-

party risks coupled with the deterioration of banks' loan

portfolio. Vo (2014) utilized coexceedance approach to

examine financial contagion in Euro area and South

Asian markets using a framework of multinomial logit

regression model and daily data spanning the period

January 2007 to March 2013. Exceedances are defined as

extreme negative returns that are below a certain thresh-

old (i.e., 5% bottom tail) in one country, whereas

coexceedances refer to the joint occurrences of

exceedances in two or more markets. The study docu-

mented evidence of coexceedances during global finan-

cial crisis and Eurozone crisis. Choudhry and Jayasekera

(2015) reported that during the turbulent period of global

financial crisis, betas increased for most firms in the

United Kingdom from the precrisis to the crisis period,

and the level of market efficiency declined significantly

from the precrisis to crisis period. Asgharian,

Christiansen, and Hou (2017) examined the long-run and

short-run components of factor betas using a framework

of component GARCH model. They have applied an aug-

mented Fama–French asset pricing model to industry

portfolios using market, SMB, and HML as risk factors.

The study reported that the cross-sectional average and

dispersion of the short-run component of betas increase

in bad states of the economy. Given the anecdotal evi-

dence of significant deterioration of systematic risk and

market risk, the findings from these studies indicate that

the extent of comovement in stock markets points to ten-

dencies of the degree to which the global financial system

is perceived to be tied to common factors. Consequently,

CAPM and international CAPM (ICAPM) provide an

appropriate methodological framework to approximate

the heightened systematic risk underlying the deteriora-

tion of common factor in such turbulent market

conditions.

FIGURE 1 Selected counterparty

exposures to AIG at the time of its failure.

Sources: American International Group (d) and

Capital IQ. (a) The chart shows collateral that

AIG returned between 16 September and

31 December 2008 to retire CDS obligations

which existed at the time of its failure.

(b) Selected counterparties shown. Does not

represent total exposure to AIG. (c) Tier 1 capital

as of 30 June 2008 as reported in each bank's

accounts. Goldman Sachs data are for

29 August 2008

2ALEXANDRIDIS AND HASAN

29 August 2008 [Colour figure can be viewed at

wileyonlinelibrary.com]

ALEXANDRIDIS AND HASA 519