WP 11-16

N. Apergis
University of Piraeus, Greece

E. Mamatzakis
University of Piraeus, Greece
The Rimini Centre for Economic Analysis (RCEA), Italy

C. Staikuras
Athens University of Economics and Business, Greece

The Greek sovereign debt crisis: testing for regime changes.

N. Apergis1, E. Mamatzakis2 and C. Staikuras3

December 2010

Abstract

This  paper  examines  whether  the  efficiency  market  hypothesis  for  the  Greek sovereign debt holds. As in Blanco et al. (2005) we test the theoretical equivalence of credit  default  swap  (CDS)  and  spreads  that  dictates  a  cointegration  relationship between the two. The main innovation of the present analysis is the use of a threshold vector error-correction (TVECM) model, thus allowing thresholds within the sample covering the period 1990-2010. Moreover, by employing this  methodology we are able to evaluate the degree and dynamics of transaction costs resulting from various events due to external market imperfections but also domestic factors. The main hypothesis we test is to what extent spreads and CDS are indeed integrated that may result in an efficient and integrated segniorage capital market. Our findings support the gradual integration  hypothesis. We find that spreads and CDS are cointegrated, though threshold effects are also revealed in terms of events that have impacted on markets.

1    Department of Banking and Financial Management, University of Piraeus, Karaoli and
Dimitriou 80, Piraeus 18534, Greece, napergis@unipi.gr.
2  University of Piraeus, Department of Economics, tzakis@unipi.gr.
3Member of Greek Parliament and Department of Accounting and Finance, Athens University of Economics and Business, 76 Patision Street, 104 34, Athens, Greece, cstaik@aueb.gr .

Box1. The Sovereign Debt Crisis in Greece: Chronology of Events.
23 December 2009: Parliament adopts the 2010 budget setting a general government deficit target of 9.1 percent of GDP.
15  January  2010:  Government  submits  the  updated  stability  programme  (SP), projecting a reduction of the government deficit of 4 percentage points to 8.7 percent of GDP in 2010, and correction of the excessive deficit by 2012. The debt ratio was projected to peak at 121 percent of GDP in 2011.
1 February 2010: 2-year bond spreads reach 347 basis points; 10-years bond spreads reach 270 basis points.
3 February 2010: The Commission adopts (i) a proposal for a Council Decision, in view  of  the  excessive deficit  correction in  Greece  by 2012, (ii)  a  draft  Council Recommendation with a view to ending the inconsistency with the broad guidelines of the economic policies, and (iii) a draft Council Opinion on the SP.
2 February 2010: Greece announces a set of measures in addition to those announced in the SP (freezing wages and raising excises with the aim of reducing the government deficit).
16 February 2010: Council adopts the above-mentioned documents, after discussion in the Eurogroup.
8 April 2010: 2-year bond spreads reach 652 basis points; 10-years bond spreads reach 430 basis points.
15 April 2010: Greece requests ‘discussions with the European Commission, the ECB and the IMF on  a multi-year programme of economic policies (…) that could be supported with financial assistance, if the Greek authorities were to decide to request such assistance.’
23 April  2010: Greece  requests financial  assistance from  the  euro-area  Member
States and the IMF.

27 April 2010: 2-year bond spreads reach 1552 basis points; 10-years bond spreads reach 755 basis points.
2 May 2010: Greece, the Commission, the ECB and IMF announce an agreement on a three-year    programme    of    economic    and    financial    policies.    The    Eurogroup unanimously agrees to activate stability support to Greece via bilateral loans centrally pooled by the European Commission
6 May 2010: The Greek Parliament votes to accept a series of policy measures included in the programme of economic and financial policies, including an increase in VAT and excises, as well as further reductions in public sector wages and pensions.
6 May 2010: ECB adopts temporary measures relating to the eligibility of marketable debt instruments issued or guaranteed by the Greek Government.
7 May 2010: 2-year bond spreads reach 1739 basis points; 10-years bond spreads reach 1287 basis points.
7 May 2010: The Council adopts a Decision according to Articles 126(9) and 136 of the Treaty including the main conditions to be respected by Greece in the context of the financial assistance programme.
9 May 2010: IMF executive board approves the Stand-by arrangement (SBA).
9 and 10 May 2010: The Council and the EU Member States endorse a financial stabilisation mechanism.
18 May 2010: The euro-area Member States disburse the first instalment (EUR 14.5 billion) of a pooled loan to Greece.
Source: EU Commission.

5. Conclusions and Policy Implications

Over the past year, euro area sovereign yields have exhibited an unprecedented degree of  volatility.  In  March  2009  the  spread  between  the  yield  on  a  10-year  Greek government bond and the yield on a German Bund of equivalent maturity was as high as 280 basis points (bp). By September 2009 the same spread had dropped below 120 bp. In January 2010, it had climbed back up to over 380 bp.  Alas, in April 2010 the spread reach 670 bp only to climb even higher to the level of 1287 bp in May 2010, the   month   that   Emergency   Financing   Mechanism   and   the   memorandum   of

understanding regarding policy conditionality, a joint initiative of the IMF, the EU Commission and the ECB were signed.

The goal of this paper was to examine whether there is a cointegration relationship between CDS  and spreads of Greek sovereign bonds. We employed the threshold cointegration analysis of Hansen and Seo (2002). This methodology allows for regime shifts over the sample period and by doing so it would give us an opportunity to show whether the arbitrage relationship expected by the literature is subject to thresholds effects. The regime shift could occur in the intercept, trend or the entire cointegration vector. The main hypothesis we tested was to what extent spreads and CDS were indeed integrated that may result in an efficient and integrated segniorage capital market. The  TVECM model has the advantage of allowing thresholds within the sample covering the period 2003-2010.

Our findings support the gradual integration hypothesis. Moreover, we found that spreads over swaps and CDS were cointegrated and two regimes were identified. This implies that a threshold exists and it is indeed significant. Thus, adjustment costs in the error correction are present, and they are valid at the typical regime one that is the dominant, and as a result should not be ignored.

Η συνέχεια εδώ>