Discussion Paper

No. 2014-9 | February 25, 2014
The Causal Linkages between Sovereign CDS Prices for the BRICS and Major European Economies

Abstract

The article examines causal relationships between sovereign credit default swaps (CDS) prices for the BRICS and most important EU economies (Germany, France, the UK, Italy, Spain) during the European debt crisis. The cross-correlation function (CCF) approach used in the research distinguishes between causality-in-mean and causality-in-variance. In both causality dimensions, the BRICS CDS prices tend to Granger cause those of the EU counterparts with the exception of Germany. Italy and Spain exhibit the highest dependence on the BRICS, whereas only India has a negative balance of outgoing and incoming causal linkages among the BRICS. Thus, the paper underscores the signs of decoupling effects in the sovereign CDS market and also supports the view that the European debt crisis has so far had a limited non-EU impact in this market.

Data Set

Data sets for articles published in "Economics" are available at Dataverse. Please have a look at our repository.

The data set for this article can be found at: http://dx.doi.org/10.7910/DVN/24788

JEL Classification

C50 G10 G15

Cite As

Mikhail Stolbov (2014). The Causal Linkages between Sovereign CDS Prices for the BRICS and Major European Economies. Economics Discussion Papers, No 2014-9, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2014-9

Assessment



Comments and Questions


Tongam Sihol Nababan - Introduction
February 26, 2014 - 01:56

I think in subject of Introduction, you should explain why you choose to compare between Major European Economic and BRICS. It is important to convince us the two groups are different in economic growth rate. So, it needs more explanations in your background of research.


Mikhail Stolbov - Reply to the comment
February 26, 2014 - 08:23

Thank you for the comment concerning the introduction of my paper. My focus is on financial spillovers between the BRICS and key EU economies. The difference in growth rates is, by all means, important but it should be studied separately rather than in the context of financial causality. Had I ...[more]

... focused on GDP growth rates in the introduction, I would have misled readers.


Massimiliano Caporin - Comments
February 26, 2014 - 09:59

I think the topic you propose is interesting. However, I have the following remarks:

- is there any possibility of extending backward the sample? It would be informative for periods where crisis were not present; moreover, are there any changes in the relation across countries after the start of ...[more]

... ECB LTRO?
- equations 1 and 2 cannot be defined as standardizations of GARCH residuals, but should be set to standardizations of the original series by using the conditional mean from the ARMA part of the model and the conditional variance from the GARCH part of the model;
- I have some doubts on the ARMA-EGARCH specifications for Brazil and UK; in fact, by looking at the coefficients, it seems to me that the AR and MA polynomials cancel out; why not testing by likelihood ratio the best-specification you identify with a specification without ARMA terms? This is viable since the models will be nested;
- I have some doubts on the use of daily data; in fact, at the daily level the EU markets are only partially overlapping with Brazil, India, and not overlapping with China; this would imply that there is a contemporaneous information flow from China to EU which you do not take into account, and a contemporaneous information flow from India to EU and from EU to Brazil; I'm not sure if those missing linkages are in practice hiding or distorting the relations across countries; I would suggest a double check with weekly data.


Mikhail Stolbov - Reply to the comment
February 26, 2014 - 13:38

Thank you for the comment. I intentionally focused on the examination of causal relationships between the BRICS and the EU in terms of the crisis. It is in line with the previous studies based on CCF, e.g. Yoshizaki et al. (2013). However, I acknowledge that it would be relevant to ...[more]

... examine the robustness of the conclusions by backtesting to at least 2008-2009. I also admit the potential usefulness of cross-check at weekly data and controlling for a possible ECB LTRO effect.


Anonymous - referee report
April 22, 2014 - 08:48

see attached file


Mikhail Stolbov - Reply to the referee report
April 23, 2014 - 10:35

see attached file