New empirical evidence on CEE's stock markets integration

• Date: 01 October 2020
• Author: Claudiu Boţoc,Sorin Gabriel Anton
• Published date: 01 October 2020
• DOI: http://doi.org/10.1111/twec.12961

World Econ. 2020;43:2785–2802. wileyonlinelibrary.com/journal/twec

|

2785

© 2020 John Wiley & Sons Ltd

Received: 23 June 2018

|

Revised: 30 December 2019

|

Accepted: 8 April 2020

DOI: 10.1111/twec.12961

ORIGINAL ARTICLE

New empirical evidence on CEE's stock markets

integration

ClaudiuBoţoc1,2

|

Sorin GabrielAnton3

1Department of Finance, Faculty of Economics and Business Administration, West University of Timișoara, Timișoara,

Romania

2Monetary Research Center, Studentski Grad “Hr. Botev”, Sofia, Bulgaria

3Finance, Money, and Public Administration Department, Faculty of Economics and Business Administration, Alexandru

Ioan Cuza University of Iasi, Iasi, Romania

KEYWORDS

cointegration, diversification, EU integration, rolling window, stock market

1

|

INTRODUCTION

Assessing the behaviour of international stock markets represents a key issue for global investors,

issuers and financial regulators in order to deal with crises spreading from one stock market to an-

other, most of the time from developed to developing economies. International portfolio diversifi-

cation seeks for potential benefits, and this determines the investors to migrate from developed stock

markets to the emerging ones. Such migration is influenced by factors like regulations, currency

volatility, financial system stability or the economic conditions. At the base of this international

diversification lies out the so-called stock market cointegration, which in the last decades emerged

from globalisation and economic integration. Financial integration represents a keen strategic goal

at EU level,1 given that “a deeper financial integration may have an impact on to the stability of the

whole financial system” (ECB,2018). In last decade, the ECB’s price-based composite indicator of

financial integration exhibit diverse tendencies. First, there was a significant deceleration trend

(2007–12) associated with the financial and sovereign debt crises between 2007 and 2011. Second,

there was a re-integration trend (2013–15) which was a consequence of European Central Bank's

Outright Monetary Transactions (OMT) framework and European banking union announcements

from 2012. Third, a temporary correction was recorded between late 2015 and end 2016, followed

by a pronounced increase during 2017. Such upward and downward trends constitute a motivation

of re-examine previous findings.

1Is one of the three objectives included in the mission statement of the Euro-system (ECB).

2786

|

BOŢOC and anTOn

When looking at Europe, the main geopolitical event is represented by the EU enlargement

process, the increasing integration of Central and Eastern European (CEE hereafter) countries

with those of developed European countries. Since the joining process requires among others

increasing capital, as well as factor and product market integration, it is likely that such structural

fundamental changes to affect the behaviour of stock markets. Due to the accession in different

waves, an important question is still open for debates among researchers and practitioners. That

is, if this association between countries lead or not to the stock market cointegration, that is

whether result in potential diversification benefits. It should be remarked that CEE’s stock mar-

kets are quite young and in their modern form (post-communism period) are dated only from 1990

onwards.

Given this, the aim of the study was to explore both the long-run and short-run integration of CEE

stock markets with those of developed ones. Using data from 20 October 2000 up to 20 October 2016,

the results from static approach suggest evidence on long-run cointegration for CEE stock markets

with Germany, UK and the USA. When the dynamic approach was employed through the rolling win-

dow estimation, either cointegration or correlation, all CEE markets analysed exhibit several short-

run cointegration relationships. This means that in specific period's portfolio diversification could be

managed through several CEE stock markets.

Since the empirical literature regarding the cointegration between CEE and developed stock mar-

kets yielded different findings that are sample dependent, in this study we extend the evidence by

refining and updating previous work, particularly during sub-periods that reflect the effects of com-

mon shocks (EU joining, financial crisis, Euro sovereign crisis). In this respect, the first contribution

is related to the recent period examined (up to October 2016) in order to reinforce or reject previous

results. A similar analysis performed by Harrison and Moore (2009) includes only the period 1994–

2006; therefore, the following events (crises) that shaped the financial integration process motivate

our study. Second, we examine both pairwise and group correlation that involves CEE markets and

each developed market, for the purpose of avoiding potential inferences among developed markets.

Third, given the superiority of dynamic cointegration, we employ three approaches in terms of meth-

odology: (a) time-varying realised correlation ratios; (b) time-varying cointegration statistics; and

(c) a multivariate GARCH model. Supplementary to the previous research (i.e. Gilmore, Lucey, &

McManus, 2008), we go beyond the largest top three CEE states—in terms of market capitalisation

(Poland, Czech Republic and Hungary) and include finally seven countries from three different re-

gions (Central, South-East and Baltic Europe). Unlike Harrison and Moore (2009) we complement

the cointegration approach with a superior multivariate GARCH model, that is asymmetric dynamic

conditional correlation model instead of Baba-Engle-Kraft-Kroner (BEKK) model. Contrary to the

previous research (i.e. Guidi & Ugur, 2014), we use daily data that seem to exhibit fewer peaks within

dynamic conditional correlation than weekly data.

The study is organised into five sections. Section2 provides an overview regarding empirical

investigations related to the cointegration of CEE stock markets. Section3 deals with the data and

methodology employed for testing the cointegration relationship. Section4 provides the empirical

results with appropriate interpretations and Section5 concludes.

2

|

LITERATURE REVIEW

The study is closely related to a various areas of literature on stock markets interactions and con-

tagion across other markets. Given the difficult applicability of standard definitions for financial