Giovanni Calice. Spillovers in Sovereign Bond and CDS Markets: An Analysis of The Eurozone Sovereign Debt Crisis
1. Introduction
Methodology
Results
Liquidity Spillovers in Sovereign Bond and CDS
Markets: An Analysis of The Eurozone Sovereign
Debt Crisis
Giovanni Calice
School of Management, University of Southampton, England, U.K.
Jing Chen
School Business and Economics, Swansea University, Wales, U.K.
Julian Williams
Business School, University of Aberdeen, Scotland, U.K.
February 2012
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
2. Introduction
Methodology
Results
Eurozone Sovereign Crisis
Ongoing issue of liquidity and solvency of various EU governments.
Causes are diverse (poor fiscal planning in Portugal, expensive bank
guarantees in Ireland, falsified national accounts in Greece)
At present Greece, Portugal and Ireland are in receipt of financial
guarantees and liquid capital injections via the IMF, EFSF and asset
purchases by the ECB.
Iceland has also received a substantial ‘bail-out’ after the collapse of
its banking system, earlier on in the crisis.
Causes are well known and are for other discussions.
This paper looks at the mechanism of transmission of liquidity and
information in the price formation mechanism of Eurozone sovereign
debt during the 2007-2011 period.
The paper provides a table of various macroeconomic indicators for
2007, 2008, 2009 and 2010 versus the 2001-2006 average.
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
3. Introduction
Methodology
Results
Outline of Talk
Brief overview of our research questions and methodology.
Our data and the uniqueness of the data set.
A short tour of some of the main results.
Brief concluding remarks.
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
4. Introduction
Methodology
Results
Liquidity and Price Formation in Crises
Variables
Let BON Dt be the yield (or discount premia) for each countries
sovereign debt (for either 5 or 10 year maturities) measured in basis
points.
CDSt is the credit default swap rate, in basis points for each
country. BON DDE,t and CDSDE,t are respectively the yield and
CDS spread on German sovereign debt of 5 and 10 year maturity.
BON DBIDt is the bid yield in basis points for sovereign bonds and
BON DASKt is the ask yield for sovereign bonds, again converted
to basis points.
CDSBIDt and CDSASKt are, respectively, the bid and ask
spreads for 5 and 10 year sovereign CDS in basis points.
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
5. Introduction
Methodology
Results
Liquidity and Price Formation in Crises
Our key research question is to establish the dynamics of interaction
between the credit spread on traded Eurozone sovereign debt with the
credit spread on equivalent maturity sovereign CDS and the liquidity
spreads on traded sovereign debt and CDSs.
For each country we compute the BON DCSt , the sovereign bond
credit spread, the CDSCSt , the CDS credit spread, the
BON DLSt , the sovereign bond liquidity spread and finally the
CDSLSt , the CDS liquidity spread. These are computed as follows:
BON DCSt = BON Dt − BON DDE,t (1)
CDSCSt = CDSt − CDSDE,t (2)
BON DLSt = BON DBIDt − BON DASKt (3)
CDSLSt = CDSBIDt − CDSASKt (4)
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
6. Introduction
Methodology
Results
A Time Varying Vector Autoregression
In this paper we provide results for an endogenous time varying VAR
model of price and liquidity formation for sovereign bond and CDS
markets during the crisis.
BON DCSt = β1,1,t BON DCSt−1 + β1,2,t CDSCSt−1
+β1,3,t BON DLSt−1 + β1,4,t CDSLSt−1 + µ1,t + u1,t
CDSCSt = β2,1,t BON DCSt−1 + β2,2,t CDSCSt−1
+β2,3,t BON DLSt−1 + β2,4,t CDSLSt−1 + µ2,t + u2,t
BON DLSt = β3,1,t BON DCSt−1 + β3,2,t CDSCSt−1
+β3,3,t BON DLSt−1 + β3,4,t CDSLSt−1 + µ3,t + u3,t
CDSLSt = β4,1,t BON DCSt−1 + β4,2,t CDSCSt−1
+β4,3,t BON DLSt−1 + β4,4,t CDSLSt−1 + µ4,t + u4,t
(5)
the coefficients [βi,j ] are collected into the time varying matrix Bt .
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
7. Introduction
Methodology
Results
We have developed a least squares based alternative to the Kalman
filter that is robust to structural change, whilst being able to capture
local stability in the coefficients.
We call this approach recursive and iteratively re-weighted least
squares (IRLS), which might be thought of as a specific class of the
extended least squares approach.
More specifically, the model is a multivariate extension of the single
equation autoregressive model of Arvastson et al. 2000 which is a
standard autoregressive model with time varying coefficients
estimated with exponential forgetting.
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
8. Introduction
Methodology
Results
˜
The eigenvalues of the time varying matrix Bt offer valuable
information on the instantaneous stability of the autoregressive
model.
Consider the time varying eigenvalues of the 4 × 4 slope matrix Bt , ˜
ordered from largest to smallest as {λmax,t , λ2,t , λ3,t , λmin,t }.
We have imposed a first order VAR on the time varying coefficients,
therefore the eigenvalues of this matrix correspond directly to
polynomial roots of the VAR process.
If the range of λmax,t to λmin,t is within the unit circle then the
instantaneous static VAR at time t is stationary. A root equal to one
indicates the presence of at least one random walk in the vector
system.
Roots greater than unity indicate an explosive stochastic trend.
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
9. Introduction
Methodology
Results
Variance breakpoint tests
Another helpful by-product of the recursive regression approach is
that a standard matrix equality test can be used to extend the
standard variance break point tests for structural breaks,
By use of a Wishart style covariance equality test, details are in the
paper.
The idea is to identify whether the conditional covariance matrix at
t is equal to the long run covariance matrix Σ from the model
residuals.
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
10. Introduction
Methodology
Results
Data set
Given the controversy surrounding the reporting of various credit
spread indices, we have constructed our data set, where possible,
from the transaction history.
The data set is sourced from Thomson-Reuters Tick History and
DataStream. Sovereign bond data is collected using the ‘Super
RICs’ or Reuters Information Codes.
The super-RICs collect all trades on instruments in the tag range set
by the code, i.e. AT5YT=RR literally means pull all yields on traded
bonds with a 5 year maturity from the daily collection date.
We use the same approach for the CDS market, however aggregation
is much more complex. Multiple data vendors provide an array of
intra-day and end-of-day information, through Markit and CMA.
The CDS data set is then hand built from these sources and
combined into a daily index.
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
11. Introduction
Methodology
Results
Countries in sample
We collect all traded sovereign bonds with a maturity of 5 and 10
years for the countries selected in the sample.
Originally all Eurozone countries were included in the sample.
However, credit default swaps have only been actively traded on ten
countries for a long enough period to permit analysis.
These countries are Austria, Belgium, France, Germany (the
benchmark), Greece, Ireland, Italy, Netherlands, Portugal and Spain.
The next slide lists the various CDS sources that CMA and Markit
use when building the index of daily spreads.
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
12. Introduction
Methodology
Results
ABN AMRO ANZ Investment Bank (Asia)
Barclays CDS NYC Barclays Tokyo
BNP Paribas Citigroup Global Mkts
Deutsche Bank NY Deutsche Bank Singapore
DZ Bank, Frankfurt GFI Market Recap
Handelsbanken Hypovereinsbank
ICAP ING Manila
J.P.Morgan Mizuho Securities
Natexis Nord LB, Hannover
RBS Japan SEB
Standard Chartered Singapore TIFFE
Tullett Prebon UBS Japan
UBS Singapore CMA
Markit
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
13. Introduction
Methodology
Results
Country Ticks Zero Yields Corrupted Rogue Trading Days
Spain 4,010,003 606 0 24 1,339
Austria 5,609,129 348 0 12 1,339
Belgium 978,395 55,981 0 0 1,339
France 708,122 31,168 0 2 1,339
Germany 2,141,828 61 0 2 1,339
Greece 2,800,111 18,574 0 4 1,339
Ireland 3,151,086 4,982 0 6 1,339
Italy 3,800,255 299,131 0 3 1,339
Netherlands 4,866,969 593 0 4 1,339
Portugal 4,616,628 10,253 0 3 1,339
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
14. Introduction
Methodology
Results
Results
Large number of results in the paper, appendix and internet
appendix.
The results are ordered in the paper as follows:
Breakpoint tests (points at which the market has appeared to change
pricing model).
Time varying roots (detecting the presence of explosive stochastic
trends, helpful for policy makers).
Time varying coefficients (direction of price discovery mechanism in
the market).
First: A visual inspection of the data for Greece, Ireland, the
Netherlands and France.
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
15. Introduction
Methodology
Results
Greek credit spreads
Credit Spreads 5 Year
1400
Bond
1200 CDS
1000
800
600
400
200
0
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
Credit Spreads 10 Year
1000
Bond
CDS
800
600
400
200
0
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
16. Introduction
Methodology
Results
Greek liquidity spreads
Liquidity Spreads 5 Year
250
Bond
CDS
200
150
100
50
0
−50
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
Liquidity Spreads 10 Year
120
Bond
CDS
100
80
60
40
20
0
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
17. Introduction
Methodology
Results
Irish credit Spreads
Credit Spreads 5 Year
500
Bond
CDS
400
300
200
100
0
−100
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
Credit Spreads 10 Year
500
Bond
CDS
400
300
200
100
0
−100
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
18. Introduction
Methodology
Results
Irish liquidity Spreads
Liquidity Spreads 5 Year
80
Bond
70 CDS
60
50
40
30
20
10
0
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
Liquidity Spreads 10 Year
40
Bond
35 CDS
30
25
20
15
10
5
0
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
19. Introduction
Methodology
Results
Dutch credit spreads
Credit Spreads 5 Year
100
Bond
CDS
80
60
40
20
0
−20
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
Credit Spreads 10 Year
120
Bond
100 CDS
80
60
40
20
0
−20
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
20. Introduction
Methodology
Results
Dutch liquidity spreads
Liquidity Spreads 5 Year
30
Bond
CDS
25
20
15
10
5
0
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
Liquidity Spreads 10 Year
35
Bond
30 CDS
25
20
15
10
5
0
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
21. Introduction
Methodology
Results
French credit spreads
Credit Spreads 5 Year
100
Bond
CDS
80
60
40
20
0
−20
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
Credit Spreads 10 Year
70
Bond
60 CDS
50
40
30
20
10
0
−10
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
22. Introduction
Methodology
Results
French liquidity spreads
Liquidity Spreads 5 Year
20
Bond
CDS
15
10
5
0
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
Liquidity Spreads 10 Year
15
Bond
CDS
10
5
0
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
23. Introduction
Methodology
Results
Detected First Variance Breakpoints
Austria (AT) Belgium (BE)
5 Year 10 Year 5 Year 10 Year
February 2008 May 2007 August 2007 May 2008
France (FR) Greece (GR)
5 Year 10 Year 5 Year 10 Year
March 2008 January 2007 January January 2007
Ireland (IE) Italy (IT)
5 Year 10 Year 5 Year 10 Year
May 2008 September 2008 March 2008 November 2009
Netherlands (NL) Portugal (PT)
5 Year 10 Year 5 Year 10 Year
March 2008 January 2007 February 2008 March 2010
Spain (ES)
5 Year 10 Year
March 2008 August 2007
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
24. Introduction
Methodology
Results
Next Few Slides
Document the time varying roots of the first order coefficients
matrix, for Greece and Portugal.
Roots above unity indicate the presence of explosive trends.
Roots equal to one indicate that there is at least one random walk in
the vector process.
In our internet appendix, we document the results for every country
and adjust the nuisance parameters in the weighting system to
illustrate the robustness of the results.
For the smallest root, if it is very large, then this indicates a jointly
explosive trend.
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
25. Introduction
Methodology
Results
5 year Greek model roots.
Roots
1.4
Largest Root
Smallest Root
1.2
1
0.8
0.6
0.4
0.2
0
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
26. Introduction
Methodology
Results
10 year Greek model roots.
Roots
1.2
Largest Root
Smallest Root
1
0.8
0.6
0.4
0.2
0
−0.2
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
27. Introduction
Methodology
Results
5 year Portuguese model roots.
Roots
1.2
Largest Root
Smallest Root
1.1
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
28. Introduction
Methodology
Results
10 year Portuguese model roots.
Roots
1.4
Largest Root
Smallest Root
1.2
1
0.8
0.6
0.4
0.2
0
Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 Q2−08 Q3−08 Q4−08 Q1−09 Q2−09 Q3−09 Q4−09 Q1−10 Q2−10 Q3−10 Q4−10
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
33. Introduction
Methodology
Results
Observations
Most important take home messages:
Explosive trends present at times in almost all Eurozone countries
and in particular Greece, Ireland and Portugal.
At this point the market has ceased to function in the normal
manner.
Without intervention the discount rate would have been driven to
infinity.
There is a time varying transmission effect from the CDS liquidity
spread to the bond market credit spread (violates the nearly
complete market condition of Jarrow-Protter 2005).
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets
34. Introduction
Methodology
Results
Policy Implication
Setting the effective rate of interest using the market rates, just
prior to bailout is inappropriate.
At this point the market has ceased to price new information and
default is already priced in, before it has happened.
This is most certainly a liquidity effect.
At points this liquidity effect is NOT from the bond market, but
from the CDS.
Which the authors believe is part of a case for banning what should
be a redundant asset.
GC,JC & JW Liquidity Spillovers in Sovereign Bond and CDS Markets