ECB'S non‐standard monetary policy and asset price volatility: Evidence from EU‐6 economies

• Date: 01 January 2021
• Author: Alessio Ciarlone,Andrea Colabella
• Published date: 01 January 2021
• DOI: http://doi.org/10.1002/ijfe.1861

RESEARCH ARTICLE

ECB'S non-standard monetary policy and asset price

volatility: Evidence from EU-6 economies

Alessio Ciarlone PhD, Senior Advisor | Andrea Colabella PhD, Advisor

Bank of Italy Economics, Statistics and

Research Directorate General,

International Relations and Economics

Directorate International, Technical

Cooperation and EU Neighbouring

Economies Division, Banca d'Italia, Rome,

Italy

Correspondence

Alessio Ciarlone, Bank of Italy

Economics, Statistics and Research

Directorate General, International

Relations and Economics Directorate

International, Technical Cooperation and

EU Neighbouring Economies Division,

Banca d'Italia, Rome, Italy.

Email: alessio.ciarlone@bancaditalia.it

Abstract

In this paper, we provide evidence that the effects of the different waves of

asset purchase programmes implemented by the ECB from 2009 onwards have

spilled over into a group of six Central and Eastern European economies

belonging to the EU but not to the euro area, contributing to shield their finan-

cial markets from the negative shocks that hit international investors' degree

of risk aversion in recent years. By means of a Dynamic Conditional Correla-

tion Multivariate Generalized Autoregressive Conditional Heteroskedasticity

Model, and resorting to three different proxies to describe the functioning, and

measure the impact, of the ECB's programmes of asset purchase, we show that

such non-standard monetary measures have had a marked role in dampening

volatility spikes in EU-6 financial markets, likely reflecting the working of a

“risk taking”and a “liquidity”channel of transmission. Results are in general

robust to an extensive series of tests, including changes in the estimation

methodology.

KEYWORDS

Central and Eastern Europe, ECB, GARCH models, international spillovers, unconventional

monetary policy, volatility

JEL CLASSIFICATION

C32; E52; E58; F3; F4; F16; F37; G1; G11; G14

1|INTRODUCTION

1

Over the last decade, central banks around the world

have embarked on an unprecedented effort to tackle the

negative consequences of the string of adverse shocks

that hit the global economy: they not only slashed refer-

ence rates to historical lows but also launched innovative

“unconventional”tools, among which the so-called

“quantitative”and “credit easing”programmes promi-

nently stood out. Like other central banks in advanced

economies (AEs), the European Central Bank (ECB) res-

orted to a series of non-standard monetary measures as

well, aimed to cope with a variety of unusual risks that

the euro area has been facing in recent years, including

liquidity disturbances in certain financial markets, fears

of a euro break-up and the ensuing “redenomination

risk”and, more recently, a prolonged period of exces-

sively low inflation. Among such non-standard measures,

asset purchase programmes (APPs) have increasingly

gained in importance and, as of late 2017, ended up

accounting for around 55% of total assets in the consoli-

dated balance sheet of the Eurosystem.

An ample literature has investigated the international

spillover effects and the related transmission channels of

such unconventional monetary policies with regard to

the experience of both the Federal Reserve (Ahmed &

Zlate, 2014; Bowman, Londono, & Sapriza, 2015; Chen,

Filardo, He, & Zhu, 2012, just to name a few) and the

Received: 15 May 2019 Revised: 8 October 2019 Accepted: 18 June 2020

DOI: 10.1002/ijfe.1861

Int J Fin Econ. 2021;26:1503–1530. wileyonlinelibrary.com/journal/ijfe © 2020 John Wiley & Sons, Ltd. 1503

ECB (Ciarlone & Colabella, 2016; Falagiarda, McQuade,

& Tirpák, 2015; Georgiadis & Gräb, 2015). These studies

mainly paid attention to the cross-border impact of

unconventional monetary policies on the level of both

financial and real variables in emerging market econo-

mies (EMEs). Only very recently the focus has begun to

shift to analyzing the consequences of such monetary

policy measures on the volatility of both real and finan-

cial variables, at home as well as abroad, hence focusing

on “volatility spillovers”in the words of Apostolou and

Beirne (2017).

The aim of this paper is to contribute to this relatively

new strand of literature. In particular, we intend to gauge

whether and to what extent the implementation of the

APPs by the ECB might have played a role in shielding

financial markets in six Central and Eastern European

economies belonging to the European Union but not to

the euro area (the EU-6) from adverse external shocks to

international investors' degree of risk aversion.

2

As these

countries are deeply integrated with the euro area

through strong financial linkages –the euro area is the

source of large capital flows towards them, while their

domestic banking systems are largely dominated by euro

area banking groups –there are good reasons to suspect

that their equity, long-term government bond and foreign

exchange markets may be subject to volatility spillovers

stemming from the ECB's APPs.

We answer to this research question by means of the

econometric approach suggested by Ananchotikul and

Zhang (2014), which is based upon the estimation of sep-

arate country-specific Dynamic Conditional Correlation

Multivariate GARCH (DCC-MGARCH) models on the

series of the levels of asset returns of the three markets at

stake. Such models have been shown to be able to use-

fully take on board asset return volatility clustering while

allowing for relationships between the volatility processes

of the asset markets under investigation which, in turn,

captures important cross-market spillover effects.

To gauge the volatility spillovers of the ECB's APPs

on EU-6 financial markets, three proxies –now rather

common in the literature –of such non-standard mone-

tary measures have been used: (a) the weekly average of

10-year yields on euro area AAA-rated government

bonds; (b) the weekly average of the shadow rate devel-

oped by Wu and Xia (2016) for the euro area; and (c) a

quantity, rather than a price, indicator represented by the

increase in the ECB's holdings of securities for monetary

purposes.

However, as the two price measures could have been

affected by several other elements in addition to the

direct impact of the ECB's APPs, we factor in this right

consideration by using two refinements. As for the euro

area AAA-rated government bond yields, we resort to the

two-stage procedure originally proposed by Ahmed and

Zlate (2014), which is intended to isolate the changes in

long-term yields that can be considered as directly attrib-

utable to the implementation of the programmes of asset

purchases. As for the shadow rate, we augment the basic

specification with a couple of additive and interaction

dummies to look at whether the fall of the shadow rate

into negative territories in December 2011 –traditionally

considered in the existing literature as a reflection of the

implementation of unconventional monetary policies –

may have altered the impact of our proxy on volatility

developments in EU-6 financial markets.

Overall, estimation results clearly point to the conclu-

sion that the implementation of the different waves of

the ECB's APPs was able to shield EU-6 financial markets

from the impact of negative external shocks to interna-

tional investors' degree of risk aversion. We relate this

taming effect to the working of a “risk-taking”and a

“liquidity”channel of transmission related to the highly

accommodative non-standard monetary policies

implemented by the ECB, originally put forward by Borio

and Zhu (2012) and empirically confirmed by Bekaert,

Hoerova, and Lo Duca (2013) and Bruno and Shin (2015).

These conclusions seem to be valid against a large series

of robustness tests based on alternative model specifica-

tions and econometric procedures. In a monetary policy

perspective, such results have important implications: in

fact, looking forward, it cannot be ruled out that the pro-

cess of gradual re-calibration of the monetary stance by

the ECB could be accompanied by an increase in volatil-

ity in EU-6 financial markets.

We contribute to the existing literature along several

dimensions. First of all, we provide new insights to the

nascent debate about the spillovers triggered by uncon-

ventional monetary measures implemented by AEs cen-

tral banks onto volatility developments in EMEs financial

markets which, to the best of our knowledge, has

appeared only in a limited number of studies up to date.

Second, we rely upon an econometric technique which is

able to adequately take on board both volatility clustering

and important cross-market correlations while allowing,

at the same time, a sufficient degree of flexibility in the

estimation procedure and a relatively light computational

burden. Third, we analyse volatility spillovers by

resorting to a wide set of proxies which are adequately

“treated”to describe, as far as possible, the actual impact

of the ECB's non-standard monetary policies. Fourth, the

estimation exercise is also performed on a rather long

time span, ranging from July 2009 to December 2016,

which enables us not only to obtain more reliable and

accurate estimates of the different linkages underlying

the chosen variables, but also to study the whole period

throughout which the ECB implemented different waves

1504 CIARLONE AND COLABELLA