Interdependence between USA, Western European and CEE Stock Markets

1
Interdependence between USA, Western European
and CEE Stock Markets
THE ACADEMY OF ECONOMIC STUDIES BUCHAREST THE
FACULTY OF FINANCE, INSURANCE, BANKING AND STOCK
EXCHANGE DOFIN - DOCTORAL
SCHOOL OF FINANCE AND BANKING

• MSc Student Elena-Manuela Tocila
• Supervisor PhD. Professor Moisa Altar

Bucharest, July 2010
2
Dissertation paper outline

• Importance of studying interdependence
• Literature Review
• Aims of the Paper
• Data
• Methodology and Results
• Conclusions
• References

3
Importance of studying interdependence

• Examining interdependence during crises is an
  important step in understanding market efficiency
  and information flows.
• Investigating interrelationships among
  international stock markets provide useful
  information for portfolio managers tight market
  linkages that appear during turmoil periods limit
  the international portfolio diversi?cation
  bene?ts and should be taken into consideration
  when deciding how to form portfolios.

4
Literature Review

• Studies about the linkages between CEE and
  developed countries
• Syrioupoulus (2004) finds that CEE markets tend
  to display stronger long-run linkages with the
  mature markets than with their neighbours.
• Serwa and Bohl (2005) analyze changes in
  interdependence for the CEE countries and finds
  that CEE markets are not more prone to contagion
  than more developed stock markets.
• Egert and Kocenda (2007) find that there are no
  long-run links between Western and Eastern stock
  markets but they conclude that there are signs of
  spillover effects (both in terms of returns and
  volatility) among CEE markets, among Western
  markets and from Western to CEE markets. However,
  no spillover seems to occur from East to West
  markets.

5
Literature Review

• Studies regarding interrelationships between
  stock markets during crises
• Forbes and Rigobon (1999) examine the
  stock-market co-movements during the following
  periods the 1997 East Asian crisis, the 1994
  Mexican peso collapse and the 1987 US stock
  market crash they find that during the three
  crises there was no contagion, only
  interdependence between the markets that
  generated the crises and those from East Asia,
  Latin America, OECD.
• Gklezakou and Mylonakis (2009) depicted that
  South European markets that are loosely related
  in periods of normal economic activity, exhibit
  strong interrelationships under conditions of the
  current economic recession. They also found that
  the USA market continued to exert dominant
  influence to the other stock markets.
• Chung et al. (2010) present the impact of
  2007-2009 global financial crisis on the
  interdependence between US market with respect to
  UK, Hong Kong, Japan, Australia, Russia and China
  markets. Their conclusion is that the
  interdependence among global stock markets become
  stronger during the crisis.

6
Aims of the Paper

• This paper aims at contributing to the analysis
  of the linkages between international stock
  markets during the current financial crisis
  through the following aspects
• to determine the extent in which the subprime
  crisis influences the developed and the emerging
  countries of Europe
• to find out how interdependence between equity
  markets could change during crisis.

7
Data

• We use the following markets in our analysis
• - US market SP 500
• - three Western European markets FTSE
  (United Kingdom), DAX (Germany) and CAC (France)
• - five CEE markets WIG 20 (Poland), PX
  (Czech Republic), BUX (Hungary), SOFIX (Bulgaria)
  and BET10 (Romania).
• Data used weekly indices in logarithm, weekly
  indices transformed into continuously compounded
  returns and the volatilities for each of the
  series (determined using GARCH models).
• Ri,t log(Pi,t-1/Pi,t), where Pi,t
  represents the index value at time t
• The two analyzed time intervals are October
  2004 June 2007(142 weeks) and July 2007- March
  2010 (142 weeks).
• The estimations and tests were performed in
  EViews 6.

8
Methodology and results

• 1. Evidence for the crisis period
• Summary
  statistics of the returns before and during the
  crisis
• 
  
• Almost all market indices returns have dropped
  significantly since July 2007, as their average
  weekly returns turned into negative returns
  during the crisis
• The volatilities during the second period are
  higher than those characterizing the stability
  period

9
Methodology and results

• 1. Evidence for the crisis period
• Pairwise correlations between SP and other
  market indices before and during the crisis
• Using the Fisher r-to-z transformation (procedure
  developed by R. A. Fisher in 1921), we calculate
  a value of z that can be applied to assess the
  significance of the difference between two
  correlation coefficients

10
Methodology and results

• 2. Testing for contagion
• Pairwise correlations between SP and other
  market indices before the crisis and during
• the entire period considered
• Comparing the correlation coefficients during the
  crisis with those calculated for the entire
  period, we find that there is no significant
  difference between them (z-test). According to
  Kristin Forbes and Roberto Rigobon (1999), this
  fact suggests that there is no contagion between
  the markets there is only a continuation of the
  strong cross-market linkages (interdependence).

11
Methodology and results

• 3. Testing for Stationarity
• Unit Root tests used Augmented Dickey Fuller
  (ADF) and Phillips Perron (PP)
• Stationarity test Kwiatkowski-Phillips-Schmidt-Sh
  in (KPSS)
• All series of indices considered in logartihm
  are non stationary and integrated of order 1 (the
  first difference is stationary)
• For the series obtained from GARCH models
  (variances) we calculate the standard deviations
  and apply the unit root tests. We find that all
  volatilities series are stationary.

12
Methodology and results

• 4. VAR models
• VAR model (returns to returns) between SP500 and
  developed countries of
• Western Europe
• Returns on the English and French markets are
  significantly influenced by developments on the
  American market during the crisis, a fact which
  cannot be validated for the period before the
  crisis
• The significant relationship between American and
  German returns before the crisis is not
  maintained for the other period, a fact which can
  be explained by the good economic results of the
  German economy that offered the needed stability
  for the local stock market. This stability
  decoupled the DAX index from the high decreases
  of the American market.

13
Methodology and results

• 4. VAR models
• VAR model (volatility to volatility) between
  SP500 and developed countries of
• Western Europe
• During the crisis, the coefficients between SP
  and the volatilities on the European developed
  stock markets become significant.
• The trend is valid even for Germany if we
  consider the interpretation of the investors
  behavior during a crisis period they will take
  rushed decisions, without any fundamental
  analysis all financial decisions will be based
  on intuition, alarming news and other fellow
  investors reactions to these. This type of
  behavior determines a high volatility on any
  market, irrespective of the listed companies
  economic status and prospects.

14
Methodology and results

• 4. VAR models
• VAR models (returns to returns) between each
  developed market and the
• group of five countries of Central and Eastern
  Europe
• The close relationship between emergent markets
  and USA (before and during the crisis) can be
  explained by observing investors behavior when
  all markets were going up (SP500 is seen like a
  benchmark), they invested in emerging markets
  then, when the crisis started in USA in July
  2007, the first thing they did was to withdraw
  their funds from developing countries, the most
  liable to confront with economical problems.
• European developed markets UK, Germany and
  France seem to have a close connection with
  Hungary in both time intervals, which we
  interpret as a proof that the Hungarian market
  has the tightest relationship with the leading
  markets on our continent, as opposed to the rest
  of the emerging economies from CEE.

15
Methodology and results

• 4. VAR models
• VAR models (volatilities to volatilities) between
  each developed market and the group of five
• countries of Central and Eastern Europe

16
Methodology and results

• 4. VAR models
• VAR models with two variables (returns to
  returns) between SP and the other indices

17
Methodology and results

• 4. VAR models
• VAR models with two variables (volatilities to
  volatilities) between SP, CEE and Western
  European countries
• In terms of volatility we observe that most of
  the coefficients become statistically significant
  during the crisis. Using Wald test we find out a
  significantly increasing spillover effect from US
  market to the markets from UK, Germany and Poland
  (marked with red in the table).

18
Methodology and results

• 5. Granger causalityWe perform Granger causality
  tests (Wald statistics) to investigate the causal
  relationships among the returns of the US and
  European markets analyzed.
• During the crisis more causal relationships can
  be identified.
• If before the crisis the US market caused only
  four markets (Hungary, Czech Republic, Poland and
  Germany), during the crisis, US market Granger
  causes all markets analyzed except for Bulgaria
  and Germany.
• In the case of Bulgaria this analysis confirms
  the previous results in which its market
  evolution does not appear to be influenced by any
  of the developed markets. This could be the
  result of its different monetary policy (the
  exchange rate peg).

19
Methodology and results

• 6. Cointegration relationships between US and
  European indices
• We conduct a Johansen test for the pairs
  (SP-other index) to
• establish if the variables are cointegrated or
  not. Then we estimate the
• VECM for the cointegrated variables.
• The three cointegration relationships are found
  only during the turmoil.

20
Methodology and results

• 7. Impulse-response analysis
• We examine the impulse-response functions
  based on VEC models estimated in the previous
  section.
• The shocks on SP causes responses on BET,
  DAX and PX indices the inverse relation is not
  sustained by the graphics. We choose two periods
  one of 4 weeks and the other of 12 weeks. For
  both intervals, it seems that BET index responds
  to SP socks in the first week and after that the
  response is decreasing.

21
Methodology and results

• 7. Impulse-response analysis
• For DAX and PX indices the responses to SP
  maintained during several
• weeks.

22
Conclusions

• In terms of returns, we conclude that
• English and French markets are significantly
  influenced by the American market during the
  crisis
• The close relationships between SP and
  developing markets are maintained during the
  crisis
• Hungarian market has the tightest relationship
  with the leading markets on our continent.
• In terms of volatility, we find that
• Although the coefficients between SP
  volatilities and volatilities on the European
  stock markets become significant during the
  crisis, there is a significantly increasing
  spillover effect only from US market to the
  markets from UK, Germany and Poland
• There is no major influence from the part of
  FTSE, DAX and CAC indices on the developing
  markets during the crisis (except Hungary and
  Poland)
• The Hungarian and the Polish markets have the
  strongest relations with developed markets in
  terms of volatility, as opposed to the Bulgarian
  one which is independent.
• Granger causality and existence of cointegartion
  relationships only during the crisis stands for
• intensified interrelationships during crisis.
• Our results are similar to those encountered by
  the recent studies. The developed stock
• exchanges are not influenced by smaller markets
  either during stability period or during crisis.
• Also, not all developed countries have influence
  on emerging markets. During the crisis US
• market has stronger linkages with the developed
  countries than with emerging ones, although
• we can remark that the relation of the American
  market with CEE countries is stronger than
• that between the Western European countries and
  CEE.

23
References

• Awokuse, O. T., A. Chopra, and D. A. Bessler
  (2009), Structural Change and International
  Stock Market Interdependence Evidence from Asian
  emerging markets, Economic Modeling, 26, 549-559
• Bonfiglioli, A. and C. Favero (2005), Explaining
  Co-Movements between Stock Markets The case of
  US and Germany, Journal of International Money
  and Finance, 24, 1299-1316
• Cheung, W., S. Fung, and T. Shin-Chuan (2010),
  Global Capital Market Interdependence and
  Spillover Effect of Credit Risk Evidence from
  2007-2009 Global Financial Crises, Applied
  Financial Economics, 201, 85-103
• Chuang, I. Y., J. R. Lu, and K. Tswei (2007),
  Interdependence of International Equity
  Variances Evidence from East Asian markets,
  Emerging Markets Review, 8, 311-327
• Egert, B. and E. Kocenda (2007), Interdependence
  between Eastern and Western Stock Markets
  Evidence from Intraday Data, Economic Systems,
  31, 184-203
• Gklezakou, T. and J. Mylonakis (2009),
  Interdependence of the Developing Stock Markets,
  Before and During the Economic Crisis The Case
  of South Europe, Journal of Money, Investment
  and Banking, 11, 70-78
• Grey, D. (2009), Financial Contagion among
  Members of the EU-8 A Cointegartion and Granger
  Causality Approach, International Journal of
  Emerging Markets, 44, 299-314
• Forbes, K. and R. Rigobone (1999), No Contagion,
  only Interdependence Measuring Stock Market
  Co-Movements, NBER Working Paper Series
• Ozdemir, Z. A. (2009), Linkages between
  International Stock Markets A Multivariate
  Long-Memory Approach, Physica A, 388, 2461-2468
• Serva, D. and M. T. Bohl (2005), Financial
  Contagion Vulnerability and Resistance A
  Comparison of European Stock Markets, Economic
  Systems, 29, 344-362
• Syriopoulus, T. and E. Roumpis (2009), Dynamic
  Correlations and Volatility Effects in the Balkan
  Equity Markets, Journal of International
  Financial Markets, Institutions and Money, 19,
  565-587

24

• Thank you!