Read the article and write a summary about International Financial Markets in Denmark. It should look like the example in the attachmentDengjun, Z. (2015). Interdependence between nordic stock markets and financial cooperation. Review of Accounting & Finance, 14(2), 172-188. Retrieved from http://ezproxy.lib.uwf.edu/login?url=https://searc…
interdependence_between_nordic.pdf
example_weekly_article_posting.docx
Unformatted Attachment Preview
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1475-7702.htm
Interdependence between
Nordic stock markets and
financial cooperation
RAF
14,2
172
Zhang Dengjun
Department of Economics, University of Stavange, Stavange, Norway
Received 25 March 2013
Revised 25 June 2014
Accepted 14 August 2014
Abstract
Purpose – This study aims to link the financial cooperation in the Nordic region and the
interdependence between the stock markets in this area. The main emphasis is placed on the evolution
of this interdependence as the financial integration was proceeding.
Design/methodology/approach – Johansen’s cointegration technique and the exponential
generalized autoregressive conditionally heteroskedastic model are applied to test the long-run and
short-run interdependences, respectively, among Nordic stock markets. In particular, the recursive
estimation approach is used to reveal the evolution of the interdependence between those markets.
Findings – The existence of two cointegrations over the sample period indicates that the markets
depend on each other to some extent. The recursive estimation of Johansen’s model further reveals that
the interdependence had been greatly improving until late 2008. The interdependence between those
markets is also confirmed convincingly by the short-term dynamics, noting that the spillover effects
between most pairs of stock volatilities are witnessed in the empirical results.
Practical implications – The findings show the dynamics of the long-run correlations between the
Nordic stock markets, which imply the intrinsic response to the process of financial market reforms, the
2008 global financial crisis and the period after the crisis. The evidenced information about
determinants of the interdependence between Nordic stock markets is sending strong signals to
investors to enhance their investment strategies.
Originality/value – Most of the existing studies have been restricted to the static long-run and/or
short-run interdependence among those markets. However, this study contributes to the literature by
investigating the dynamics of interdependence among the Nordic stock markets over time; moreover,
the evolution of the market interdependence is sketched closely to the process of the regional financial
market reforms.
Keywords Cointegration, Interdependence, EGARCH, Nordic region, Stock
Paper type Research paper
1. Introduction
The purpose of this study is to evaluate the dynamic long-run and short-run
interdependences among Nordic stock markets. This region consists of Denmark,
Norway, Sweden and Finland and has been characterized by a series of financial reforms
in the recent decades. One of the commonly asked empirical questions is whether
Review of Accounting and Finance
Vol. 14 No. 2, 2015
pp. 172-188
© Emerald Group Publishing Limited
1475-7702
DOI 10.1108/RAF-03-2013-0036
JEL classification – G15, C13
The author would like to thank Frank Asche, Hyeongwoo Kim and Hao Li for their helpful
comments. Comments from two anonymous referees are also gratefully acknowledged. The
author appreciated Linn Furuvald from Oslo Børs ASA for the OMX data and the background
information about the Norwegian stock market.
interdependence between those markets has been inherently increased following the
operational integration in this region. Booth et al. (1997, hereafter, BMT) reported
non-existence of cointegration in the Nordic stock markets despite some spillover effects
of stock prices and volatilities between these markets from 1988 to 1994. After including
two neighboring countries (Estonia and Iceland) into the model, Nielsson (2007)
provided evidence of two cointegrations between stock prices in this region from 1996 to
2006. This increased interdependence between those markets raises a legitimate doubt
that it could be a partial, if not contributing, aftermath of the deepened financial reforms.
A thorough literature review only yields one study (Hellström et al., 2013) that
particularly bridges the interdependence between Nordic stock markets and the
regional financial reforms. The study utilized a dummy variable to reflect a discrete and
one-time change in the correlation between the Nordic stock markets, which is attributed
to the creation of a uniform trading platform in the Nordic area (see more below).
However, the statistical relationship between markets may be a gradual and on-going
process (Serletis and King, 1997), and therefore, we are attempting to contribute to a
better understanding by updating and extending previous studies on interdependence
between the Nordic markets. This study is also unique since we investigate the impact
of the financial reforms on an ongoing process of interdependence between stock
markets. The empirical results have particular implications for investors and the public
listed companies in terms of cross-border diversification and liquidity gains,
respectively.
A growing number of studies have explored interdependence between stock markets.
The methods of investigating the structure of interdependence can fall roughly into two
branches. First, static models and the corresponding estimating approaches have been
widely applied to assess stock-market interdependence in terms of spillover of stock
returns and/or volatility (Koutmos, 1996; BMT, 1997; Kim et al., 2005). Second,
recognition of the nonstationary property of stock indices has motivated researchers to
evaluate the long-run convergence between national stock markets in terms of
cointegration between the level of stock prices (Kasa, 1992; Knif and Pynnönen, 1999;
Manning, 2002; Chen et al., 2002; Masih and Masih, 2004). If a time-varying
interdependent relationship between national stock markets exists as expected in the
process of financial market integration, investors need to well understand the changing
long-run equilibrium and the short-run dynamics. Accordingly, this study provides
evidence on both long-run and short-run interdependence among the four Nordic stock
markets.
In terms of long-run relationships, Johansen’s approach (Johansen, 1991) is applied.
The evidence of cointegration among stock prices implies interdependence between
those markets. Recognizing that operational integration between financial markets is a
gradual and ongoing process, we further recursively estimate Johansen’s model to
evaluate the process of statistical integration between those markets. For the short-run
analysis, information about volatility spillover is the main concern. The asymmetric
spillover effect in the equity market is broadly documented (Koutmos, 1996; BMT, 1997;
Skintzi and Refenes, 2006; Cappiello et al., 2006), and the most common approach used is
the exponential generalized autoregressive conditionally heteroskedastic (EGARCH)
model. This model can avoid imposing non-negative constraints on GARCH parameters
and hence reduce the instabilities of the parameterization. Thus, we adopt the EGARCH
model to qualify the spillover effects of volatilities and stock returns within the Nordic
Nordic stock
markets
173
RAF
14,2
174
markets. This approach also facilitates the comparability between our empirical results
and those in the study by BMT (1997), in that the same econometric approach is used.
The empirical results reveal that the long-run interdependence between Nordic
markets has improved, and the trend is roughly in accordance with the process of the
financial market integration in this region. In addition, this long-run correlation is
echoed by the short-run dynamics, noticing that volatility transmission between
pair-wise stock indices is very common during the sample period.
The structure of the paper is as follows. We, first, provide a brief overview of the
relevant literature. Then, the financial reforms in the Nordic markets are introduced and
the data description is provided. This is followed by the introduction of econometric
approaches and analysis on estimation results. The final sector presents concluding
remarks with implications.
2. Literature review
The literature review in this section has two components. First, we review the studies
that investigate the interdependence between Nordic stock markets. Second, we
examine the studies that assess the interdependence between individual Nordic markets
and other world-leading markets to help in understanding the characteristics of the
Nordic stock markets. Particular attention is paid to the econometric approaches in
terms of both long-run equilibrium and short-run dynamics.
There are two studies that particularly examine the interdependence of the Nordic
stock markets by BMT (1997) and Nielsson (2007), respectively. In the study by BMT,
there can be found no cointegration between the four stock prices in this region and tests
the short-run dynamics by applying the EGARCH model with daily data from May 1988
to June 1994. The empirical results indicate asymmetric volatility transmissions in the
markets (with the exception of Denmark) and the existence of price and volatility
spillovers for some pairs of stock markets. While there is no feedback between price
spillovers, the volatility spillover creates a feedback effect between Sweden and Finland.
Different from that study and using data extending from 1998 to 2006, Nielsson’s
empirical findings confirm the existence of two cointegrations in the Nordic market and
other two markets in the neighboring countries. This indicates more than one stochastic
trend in those markets, and hence a limited interdependence, which is also confirmed by
empirical results generated with other econometric tools in Nielsson’s (2007) study.
Researchers are also interested in the interdependence between some of the Nordic
markets. For example, the integration between Denmark and Finland receives attention
from Pynnönen and Knif (1998). They provide evidence of cointegration within data
spanning from 1920 to 1994. However, all of the shock impulses are absorbed in both
markets, and a weak Grange causality is observed in some subsamples but not the entire
sample. On the other hand, the cointegrating relationship between Sweden and Norway
is found by Knif and Pynnönen (1999) within a multivariate vector autoregression
(VAR) model inclusive of the world-leading markets and two other Nordic countries.
In the 2000s, the interdependence between the European Union (EU) stock markets
received more attention in academia. While Kim et al. (2005) apply a bivariate EGARCH
model to assess market interdependence between the European stock markets,
including Denmark, Sweden and Finland, Büttner and Hayo (2011) take advantage of a
bivariate DCC-MGARCH model to extract the dynamic conditional correlations between
the European stock markets and analyze the determinants of stock-market integration
among EU members. Bartram et al. (2007) examine the effect of the introduction of the
Euro on the interdependence between the individual stock markets and the euro-zone
regional stock-market index. Their empirical results exhibit that Sweden experienced a
slight increase in its dependence on the euro-zone market and that Denmark and
Norway do not display a clear regime shift with respect to the interdependence on the
euro-zone markets. The uniqueness of Sweden is also documented by Cappiello et al.
(2006), who find that only the Swedish stock returns have a smaller change in volatility
after a positive shock than after a negative shock, but Norway and Denmark do not.
3. Background and data
3.1 Background
During the past decade, the very first crucial step to promote cooperation between the
Nordic markets was made in September 2003. At that time, the Stockholm Stock
Exchange (OM group) and Helsinki Stock Exchange (HEX group) merged to form the
OMX group. The new stock exchange further acquired the Copenhagen Stock Exchange
in January 2005. Since the purchase of a 10 per cent stake in Oslo Børs Holding ASA in
October 2006, the OMX group has dominated the stock market operation throughout the
Nordic region. Those acquisitions lent impetus to the cooperation between the local
equity markets and the world-leading markets. On May 25, 2007, NASDAQ announced
an agreement to purchase OMX to form the NASDAQ OMX group, which then
controlled Stockholm, Helsinki, Copenhagen and four other stock exchanges. As The
New York Times (2007) reported:
The deal Friday (May 25, 2007) is the latest combination in an increasing consolidating market
fed by clients’ demands to buy and sell stocks of companies anywhere in the world, at a faster
pace and for less money.
However, the profitability of cross-border investing portfolios depends on the statistical
interdependence between regional markets, which is the main concern of this study.
The operational financial integration between the four Nordic markets was further
strengthened by other active cooperation. Since May 2005, all OMX stock exchanges,
inclusive of Oslo Børs, have used the same trading system. All the Nordic markets have
adopted the same corporate governance codes, which were introduced in 2003, 2005 and
2006, respectively. On October 2, 2006, the OMX merged the three wholly owned
exchanges (Denmark, Sweden and Finland) into a combined list traded in a virtual
exchange. In addition, on the same date, OMX also launched several other initiatives,
including the information offering and harmonized listing requirements and a
pan-regional benchmark index. The aforementioned cooperation activities are
summarized in Table I.
During the process of the ongoing financial cooperation, the four Nordic stock
markets have become more open as the foreign ownership in the public companies has
considerably increased. In 2005, the proportions of equity foreign ownership in terms of
market value were 24.1 per cent in Denmark, 37.1 per cent in Norway, 24.1 per cent in
Sweden and 50.9 per cent in Finland (Wajid et al., 2007). The corresponding numbers in
1994 were 10 per cent, 31 per cent, 26 per cent and 30 per cent, respectively (BMT, 1997).
This indicates that the two smaller markets (Denmark and Finland) have become more
open to foreign participants.
Nordic stock
markets
175
RAF
14,2
176
Another property of Nordic markets is the fast-growing stock market capitalization
and hence the changed market structure in this region. At the end of 2006, the market
values of equity capitalization were [Euro]181 billion for Denmark, [Euro]232 billion for
Norway, [Euro]467 billion for Sweden and [Euro]234 billion for Finland. The less
comprehensive counterparts at the end of 1994 were Euro 38 billion, 30 billion, 97 billion
and 31 billion, respectively. From 1994 to 2006, the proportions of Swedish and Danish
market values out of the combined regional market capitalization declined from 49.5 per
cent to 41.9 per cent and from 19.4 per cent to 16.2 per cent, respectively. However,
Norway and Finland considerably raised their shares of market capitalization of the
regional total, up from 15.3 per cent to 20.8 per cent and from 15.8 per cent to 21.1 per
cent, respectively. Sweden has been by far the leading equity market during the recent
decades.
3.1.1 Data. We study the four Nordic stock markets by using the following
well-known indices: OMX20 for Denmark, OBX for Norway, OMX30 for Sweden and
OMX25 for Finland. The observations are adjusted for public holidays to ensure
comparability of the data from different markets. The time period of the analysis
extends from September 2001 to December 2011. The starting date is chosen to gather
consistent data because it is the date on which OMX25 (then HEX25) replaced the
previous index (FOX) and the calculation of the index was modified as well. The early
period can also serve as a base to evaluate the effect of the ongoing financial reforms on
the market interdependence.
The higher-order VAR models are more likely to capture the presence of any
far-horizon mean reversion in stock prices (Kasa, 1992). In this study, the cointegration
among Nordic markets is examined by using the time series of weekly indices. The
lower-frequency data may, however, mask the volatility transmission due to the time
aggregation and compensation effect (Arouri et al., 2011). So, the daily data are used to
evaluate the volatility transmission between different markets. Data descriptions are
illustrated in Figure 1 for the daily level data and Table II for both the logarithmic
weekly indices and daily stock return data.
As shown in Figure 1, the six vertical lines in the chart represent the events of
financial integration in the Nordic markets and the starting and ending dates of the 2008
economic crisis (the vertical broken lines), which are approximated by the highest and
lowest points in the Swedish market (September 16, 2007 and November 21, 2008,
respectively). The dates on which those events occurred are labeled in the axis. This
Date
Events
September 3, 2003
The OMX group was founded as a result of a merger between Stockholm and
Helsinki stock exchanges
The same corporate governance codes were applied in the Nordic stock markets
The OMX group acquired the Copenhagen Stock Exchange
The OMX group purchased 10 per cent share of the Oslo Stock Exchange
A combined list of Denmark, Sweden and Finland’s stock exchanges were
traded in a virtual exchange; the harmonized listing requirements were
launched
NASDAQ agreed to buy OMX
The trade was complete
2003, 2005, 2006
January 1, 2005
October 2, 2006
Table I.
Events of financial
integration in the
May 25, 2007
Nordic stock markets February 27, 2008
Nordic stock
markets
177
Notes: The data are recalculated by taking Sep 2001 to Dec 200 as the base (100);
the vertical full lines represent the important events of financial reforms in the
Nordic market (Table I); the vertical broken lines range the period of the 2008
global finance crisis, which are approximately represented by the highest and
lowest points of Sweden OMX25 during that period
setting for subsamples is done in all charts in the subsequent sections. Figure 1 indicates
that the four indices seem to move along with each other during the whole sample
period, although Norway has a high level of fluctuations over time. Before September
2003, the four indices were more volatile. Since then, the indices kept a roughly rising
trend, which was broken by the 2008 global financial crisis (the period between the two
vertical broken lines) and recovered at the end of the crisis period. Those properties
intuitively imply some commonality was underlying the stock market movements.
Table II demonstrates the statistical descriptions of the logarithmic indices of weekly
data and the daily returns data. As discussed earlier, while the weekly level data are
used in the long-run cointegration model, the daily return data are applied to the
short-run dynamics model. It is noteworthy to point out that the measures for skewness
and excess kurtosis indicate non-normal distribution for all return series and hence the
Figure 1.
Nordic stock markets
indices (daily data
from Sep 2001 to Dec
2011)
RAF
14,2
178
Variable
Mean
Logarithmic indices (weekly)
logOMX20
5.7293
logOBX
5.4076
logOMX25
6.6618
logOMX30
7.5204
Maximum
6.2406
6.1187
7.1686
8.1143
Stock returns (daily)
100⌬logOMX20
0.0126
9.4693
100⌬logOBX
0.0346
11.019
Table II.
100⌬logOMX25
0.0088
9.8650
Descriptive statistics 100⌬logOMX30
0.0131
9.2855
of indices (2001:20092011:2012)
Note: Ex. Kurt. ⫽ Excess Kurtosis
Minimum
SD
Skewness
Ex. Kurt.
4.3814
3.6777
5.3010
6.2440
0.3161
0.4927
0.3042
0.3258
⫺0.6599
⫺0.5996
⫺0.7538
⫺0.4073
0.4275
⫺0.6139
0.6552
0.1515
⫺11.723
⫺11.279
⫺8.5269
⫺8.9054
1.4106
1.7923
1.6375
1.5383
⫺0.2343
⫺0.5715
0.1061
⫺0.0018
5.5719
5.6934
3.5300
3.5196
presence of the ARCH effect. The same properties of those price series in a different
period have been documented by BMT.
4 …
Purchase answer to see full
attachment
