Timing Does Matter: Institutional Flaws and the European Debt Crisis

ABSTRACT

Financial crises are complex phenomena. Yet, at the cost of some simplifications, the literature has identified two main elements causing the crises: macroeconomic imbalances and institutional design flaws. Economists recognise that both play some role, but disagree on their strength. Using Bai and Perron’s technique and an EGARCH model we contribute to the debate on the European debt crisis by identifying break dates and changes in volatility in daily values of 10-year public bond interest rates for Greece, Italy and Spain. The results are then related to key political and institutional events. The results of our econometric exercise uncover the following facts: the crisis began in May 2010; worsened after summer 2011, as the European authorities hastened to restructure the Greek debt; improved during summer 2012, when the ECB approved the OMTs, a new programme for the purchase of bonds. On the whole, the results are compatible with an interpretation of the crisis that considers the institutional flaws as the main cause.

KEYWORDS:

• European debt crisis
• interest rates
• public debt
• event study

JEL CLASSIFICATIONS:

• G12
• G14
• H63

Acknowledgments

The authors thank Charles Wyplosz and the referees of this journal for comments leading to the improvement of the paper. The usual caveats apply.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1 To put the 2015 Five Presidents’ Report into action the European Commission introduced the European Semester, which ‘enhanced democratic dialogue, and further improved economic governance through the introduction of National Productivity Boards and an advisory European Fiscal Board which is already up and running’ (Document of the European Commission titled ‘The EU’s economic governance explained’, p. 7, downloadable at https://ec.europa.eu/info/sites/info/files/eu-economic-governance-explained.pdf).

2 The analysis of the working of this new approach, which is still a debated issue, goes beyond the scope of this paper.

3 The literature has proposed three generations of models. The first focussed on the role of external macroeconomic imbalances (Salant and Henderson 1978; Krugman 1979). They provided the theoretical background from which the International Monetary Fund drew its policies for the Latin American debt crises of the 1980s. The second focussed on the possibility that, owing to problems of the institutional organisation, the authorities are unable to control speculative movements (Obstfeld 1994; Kaminsky and Reinhart 1999; Goldfajn and Valdés 1997; Flood and Marion 1999; Chang and Velasco 1999; Sarno and Taylor 2003; Buiter 2007). The events concerning the speculative attacks against the European Monetary System in 1992 and the Mexican and the Asiatic crises in 1994 and 1997 stimulated the elaboration of this generation of models. The third generation of models underlines the role of news that change the views of financial operators on the quality of some assets (Kaminsky and Schmukler 2002; Kaminsky, Reinhart, and Végh 2003; Kaminsky, Mati, and Choueiri 2009).

4 Alessandrini et al. (2014) include the divide between productivity growth and the fiscal irresponsibility of some governments among the factors causing macroeconomic imbalances. Pérez-Caldentey and Vernengo (2012) and Hein (2013) consider that the export-led growth models adopted by the euro-countries is the main cause of macroeconomic imbalances. For Papadimitriou and Wray (2012), the macroeconomic imbalances are due to the working of the unregulated banking sector, which induced a vast expansion of debt in the peripheral euro-countries. All these authors claim that the imbalances generated deficits in the current accounts of the balance of payments of the peripheral euro-countries. After the crisis of 2007–2008 financial markets did not fund these deficits anymore, causing changes in the interest rates.

5 For an account of the debate on Cesaratto’s position among post Keynesian economists, see Febrero, Uxó, and Bermejo (2018). These authors too oppose two alternative views, one in terms of macroeconomic imbalances and the other in terms of institutional flaws.

6 They recognise that in the case of Ireland and Spain the problems were generated by a high private debt because public debt had limited dimension.

7 For a review of the literature on these flaws before 2007, see Panico and Vazquez Suárez (2008).

8 Art. 105.2, which attributes to the Eurosystem the task of guaranteeing the smooth working of the transmission mechanism of monetary policy, can come in contradiction with Art. 123 that forbids the direct purchase of sovereign bonds and the bailout of national governments. Art. 125.1, which claims that euro countries cannot guarantee the sovereign debt of another member state, can contradict Art. 122.2, which states that countries threatened by severe difficulties can receive assistance by other Union members.

9 The Appendix presents the details of the methods of this section.

10 Data were downloaded from the website of Trading Economics on 18 July 2014.

11 In this context, a negative shock on bond rates means they increase causing higher cost of debt. The opposite meaning applies to the case of a positive shock.

12 See Figure A1 in the Appendix.

13 See Figures A2 and A3 in the Appendix.

14 In December 2008, the European Council adopted the European Economic Recovery Plan, a programme of national budgetary stimulus packages. In Fall 2009, the ECOFIN approved Excessive Deficit Procedures against eight countries and revised the recommendations and the time schedules for other countries already submitted to these procedures (ECB 2009).

15 See Ardagna and Caselli (2014, 307–311).

16 https://www.consilium.europa.eu/media/25673/20100502-eurogroup_statement_greece.pdf accessed on 23 March 2018.

17 It was ‘extraordinary’ also because the ECOFIN never meets during weekends because of the high organising costs.

18 For Ardagna and Caselli (2014), some euro-countries decided to provide financial assistance to Greece to favour their own banks, which had invested in Greek sovereigns. The Treaties’ statement that the European authorities must adopt a supranational perspective when making decisions did not play a significant role.

19 For a description of these events, see Panico and Purificato (2013).

20 In the Netherlands, general elections had to be held in September. In France, in the face of the presidential elections of July 2012, President Sarkozy, with the support of German Chancellor Merkel, was trying to make up for the loss of consensus by declaring that the private sectors responsible for the crisis had to pay for the damage caused.

21 President Trichet warned about the consequence of partial default, stating that this event would have disrupted the debt markets (ECB 2011a, 2011b).

22 The PSI had a negative impact on banks’ balance sheets (ECB 2011c, p. 43) and on the transmission mechanism of monetary policy (ECB 2012a, p. 59).

23 The share of the Italian sovereign debt held by foreign investors fell from 46.2 per cent in June 2011 to 35.8 in June 2012. Spain’s fell from 39.3 per cent to 29.2. In June 2011, the Italian TARGET balance was a positive figure equal to 6 bn. euro. In June 2012, it was -274.3. The Spanish ones were -45.4 and -408.4 bn. Euros, respectively.

24 Blanchard and Leigh (2013) found that, during the early years of the European debt crisis, fiscal consolidation was negatively related to growth. For them, forecasters underestimated the size of the fiscal multipliers.

25 The plan raised some capital requirements and imposed a buffer of resources to absorb losses coming from the holding of sovereign debt. Moreover, it allowed the banks to ask the financial support of the EFSF in case of necessity (http://www.consilium.europa.eu/uedocs/cms_data/docs/pressdata/en/ec/125644.pdf, accessed on 23 March 2018). The new measures forced five Spanish banks to obtain additional capital.

26 In May 2010, following the ECOFIN’s requests, the Spanish government adopted further consolidation measures (ECB 2010). In September, the Parliament approved a reform of the labour market and a pension system in August 2011. On 27 September, it reformed the Constitution to strengthen the sound operation of the government budget.

27 In November 2011, the ECOFIN had approved new rules on government budget supervision, reinforcing the Stability and Growth Pact, and on 2 March 2012 the Council of Europe signed the Treaty on Stability, Coordination and Governance, known as the Fiscal Compact. As to the banking sector, the ECOFIN had approved on 26 October 2011 a plan to recapitalise the European banks. As a consequence, the Spanish authorities imposed new provisions on the banks (RDL 2/2012), raising their needs for new funding up to 50 billion euros (ECB 2012b).

28 To facilitate its access to international markets, the Spanish banking system had been restructured in 2010 by reducing the number of small financial institutions from 45 to 18. This process led to the practical disappearance of the ‘Cajas de ahorro’ (Saving Banks). Moreover, legislation raised capital requirements foreseeing some support from public capital.

29 Bankia had been set up in 2010 to foster the integration of the ‘Cajas de ahorro’. With the support of public capital, it rapidly became the fourth largest Spanish financial institution. In 2011, it had access to the Stock Exchange. The access was, however, based on untrue accounts, which banking supervision failed to uncover. When it was nationalised in May 2012, its 2011 accounts, which had shown a profit of 309 million euros had to be corrected to exhibit a loss of 4,369 million euros (Bergés and Ontiveros 2013, 108).

30 On May 25, after reformulating its accounts, Bankia asked the government to provide additional 19 bn. euros for the required recapitalisation.

31 The optimization problem just described implies complex computations that Bai and Perron simplify by proposing a sequential algorithm based on dynamic programming. We performed all computations using the package strucchange, version 1.5-0 (Zeileis et al. 2002) that is a component of the R system for statistical computing.