International political economy of the Irish sovereign debt crisis
The international political economy dimension was important in shaping the Irish sovereign debt crisis. In many respects, the conflict between domestic and external interests was a defining feature of the Irish crisis as a whole, and throughout the period the external dimension could scarcely be disentangled from the domestic political dynamic. On the one hand, the Irish government's decision to guarantee bank debt as well as bank deposits in September 2008, and generally to provide unfettered support for the banking system, engaged domestic sovereign resources for the benefit not only of its own citizens, but also (and arguably even more) of foreign investors (who held the majority of bank bonds), the rest of Europe and the international community (as an Irish bank implosion or default could have caused contagion to the rest of Europe and chain reactions at the global level).27 On the other hand, in November 2010 pressure from European partners, the ECB and the international community to accept external support reversed the natural debtor/creditor dynamic, with creditor countries actually pushing a debtor country to accept their resources in order to prevent a possible intensification of the crisis and ramifications at the European and global levels.28 The opposite had happened in the Greek episode, as EMU partners, and particularly Germany, tried up to the last minute to avoid engaging their resources. As a result, the EU/IMF rescue package was made available to Ireland relatively quickly, contributing to a faster resolution of the most acute phase of the crisis, and limiting the impact on financial markets in the rest of Europe and at the global level.
In that context, the ECB in particular emerges as a key de facto external veto player. The ECB played an important role in persuading Irish policymakers to request external help earlier than they would have wished; Central Bank Governor Honahan was the first Irish policymaker to concede that the country might request external aid, immediately after his participation in the ECB Governing Council meeting on November 18, 2010.29 In November, the Irish sovereign had sufficient cash reserves to cover its liquidity needs until about the middle of 2011, and Irish leaders had a preference for staying out of an external rescue programme for as long as possible in order to protect the country's economic sovereignty. Meanwhile, the ECB wanted Ireland to access EFSF/IMF funding as soon as possible, since it was concerned about the consequences of prolonged Irish turmoil for Eurozone and global financial stability - particularly with regard to contagion risk - as well as for its own balance sheet exposure.30 Looking at events in the ten days preceding the agreement on the Irish aid package at the end of November, the G20 meeting on November 12, where talks were held on the Irish crisis, marked the end of the most acute phase of the crisis, as private sector loss participation plans were watered down and the door was opened to the possibility of an external support package for Ireland. On the first and crucial issue, the finance ministers of Germany, France, Italy and Spain released a communique, clarifying that the "potential private sector involvement" would not apply to outstanding debt or be a condition of EFSF activation, that it would probably not imply outright "haircuts", and that the new mechanism would not come into effect before mid-2013 (Communique by Finance Ministers of Germany, France, Italy and Spain, 2010). On the second issue, the press reported that European policymakers, and particularly the ECB, were asking Ireland to accept emergency aid, even as Irish leaders denied the country's interest in doing so.31 As a result, market sentiment started to stabilise and Irish bond yields fell by 100 basis points in two days on November 12 and 13. However, yields failed to move below 8% and remained elevated in the following few weeks, as clarity on the overall outcome of the Irish situation remained low. It took until November 21 for the Irish government to apply for external help, and up to the end of November for the aid plan to be drafted. Throughout this period, markets remained uneasy with Irish economic fundamentals, as well with the interaction between Irish and European political developments. The period leading to the aid request on November 21 was dominated by talks between Irish and European policymakers about the need for Ireland to tap the EFSF, as well as some behind-the-scenes preparations for a possible move in that direction (Bloomberg News, November 15, 2010). The meeting of the ECB Governing Council on November 18 appeared to trigger a change in the tone of the Irish authorities with regard to a possible rescue package: Irish Central Bank Governor Honahan, a member of the ECB Governing Council, was the first Irish policymaker to concede that the country's sovereign might request an official loan on the same day (Bloomberg News, November 18, 2010). On November 21, the aid application was finally put forward (Bloomberg News, November 21, 2010), while the details of the package were hammered down in the following few days and announced on November 28 (Ireland Department of Finance, 2010d).
Notably, the ECB had both an interest in getting Ireland's crisis resolved as soon as possible and the influence to get its preferences implemented, making it a crucial holder of veto power in the default/consolidation/bail- out decision. This de facto veto power came principally from the ECB and Eurozone national central banks (Eurosystem)'s position as main supplier of liquidity for the Irish banking system. With Irish banks dependent on Eurosystem financing to survive, an ECB threat to "pull the plug" would have been equivalent to a death sentence for both Irish banks and the Irish sovereign. Indeed, besides the ECB's broader mandate concerning monetary and financial stability in the Eurozone, since the beginning of the financial crisis the Eurosystem had increasingly taken the role of liquidity lifeline for Eurozone banks.32 Moreover, since May 2010, it had started intervening directly in the secondary bond markets of the small peripheral economies through the Securities Market Programme (SMP). To the extent that the importance of these support measures grew, the ECB both acquired additional country exposure and increased its power to influence the decisions of debtor governments.33 Although the late September bank refinancing bump described in Section 5.2 had been overcome, by November 2010, Irish bank borrowing at the Eurosystem through repo operations had reached 136.4 billion euros, an amount disproportionally large relative to the country's GDP, and second only to Greece relative to bank assets.34 The lending was collateralised, but the quality of Irish collateral was deteriorating by the day. Moreover, the Irish Central Bank was acquiring significant additional exposure to banks through emergency lending assistance (ELA) operations.35 This lending was even riskier than monetary policy operations, as it was carried out with those banks unable to provide suitable collateral to the ECB window. The risks and decisions rested with the national central bank, but the ECB was clearly concerned about unilateral money creation by a national central bank and the risks that the undercapitalisation of the Irish Central Bank might create for the Eurosystem as a whole. Thus, by October 2010 the ECB Governing Council had at least three reasons for preferring Ireland to access EFSF and IMF resources to fund sovereign financing needs and recapitalise banks. First, it had an increasing risk exposure to Irish assets - some of dubious quality - acquired as collateral in open market operations; second, the excessive central bank balance sheet expansion and the impact of excessive money creation created concerns, particularly since the economy had started improving in core European economies; third, there was a significant risk that an intensification in the Irish sovereign and banking crisis would affect the sovereigns and banking systems in other EMU countries. The counterpart of the Eurosystem exposure to Irish risk was an increased influence on the decisions of Irish policymakers. An early exit from the exceptional bank liquidity support measures introduced from 2007, such as full allotment in auction and longer-term refinancing operations, or a policy rate increase, would have hurt Irish banks and the Irish economy more than most in the region. Broadly, the ECB's role in managing the price, length and amount of refinancing operations gave it a power of "life-or-death" over Irish banks and, therefore, over the creditworthiness of the Irish sovereign. Similarly, with the SMP programme, the ECB retained a choice of whether or not to intervene directly to contain moves in sovereign bond prices through secondary market purchases, thus supporting or not supporting Irish government actions. Data on weekly SMP purchases36 show a pick-up in bond purchases at the most intense phases of the crisis, but with clearer conviction once the bail-out was agreed.
While in the case of Greece we identified Germany, a sovereign partner within the EMU, as holding the key external veto, in the case of Ireland a supranational institution played that role. The evolution of the Eurozone institutional framework during the period between the two episodes,37 reducing the role of individual country creditors and empowering supranational institutions, and the importance of the banking crisis in the Irish case, contributed to making the ECB the crucial external interlocutor for Ireland as well as the key external veto player monitored by financial markets. Remarkably, Germany, in the Greek episode, and the ECB, in the Irish episode, had two crucial features in common: first, they were both external creditors;38 and, second, they both held power and resources that could prove determinant for the fate of the debtor sovereign. Since the ECB is a collective decision-making institution, where representatives from the member country central banks are represented, the issue arises as to whether in our analysis the ECB should be considered as an actor with its own independent preferences or be treated just as a reflection of the preferences of one or more member countries (Germany in particular). Our approach of analysing it as an actor with its own preferences and powers is consistent with the concept of ECB independence, a key principle of the Maastricht Treaty: members of the Governing Council are supposed to act on the basis of regional rather than specific countries' interests. In reality, it cannot be excluded that ECB Council decisions may have been directly or indirectly influenced by the preferences of some individual member countries, for example weak EMU peripheral countries fearing contagion or Germany fearing excessive money creation or credit risk exposure by the ECB.39
Going back to the role of the ECB in the Irish sovereign debt crisis, hypothesis 3.1 of our theoretical framework anticipates that, as a de facto external veto player in the default/consolidation/bail-out decision, its preferences will have been significant for investor assessment of sovereign risk. The role of the preferences and actions of the ECB in influencing the trajectory of the Irish sovereign debt crisis can be inferred from both the sequence of events and the incentives and inter-linkages highlighted above. Some specific examples as to the bond spread impact of ECB Irish bond purchases in the context of the SMP programme also corroborate this view. During the period under consideration, the impact of SMP purchases on bond spreads can be discerned in a few instances, although always within the context of a complex set of drivers. Indeed, as mentioned above, SMP purchases appeared to be timed to reward, and to reinforce the impact of, the Irish government's policy announcements as well as the acceptance of the rescue package. Thus, the step-up in bond purchases in the week ending October 1 contributed to the easing of market stress in early October, in parallel with the Irish government's policy announcements of September 29; similarly, the step-up in SMP purchases in late November and early December contributed, along with the effect of the Irish rescue deal, to arresting the increase in bond spreads. Overall, the role played by the ECB in shaping the dynamic of the sovereign debt crisis in general, and government bond spreads in particular, during the Irish sovereign debt crisis reinforces the evidence described in relation to the Greek case, where German preferences emerged as important in shaping the dynamic of sovereign bond spreads.
Hypothesis 3.2 of the theoretical framework takes the analysis a step further, postulating that a higher degree of proximity between debtor and creditor countries will increase the debtor's credibility in financial markets. In contrast to Greece's low level of European and international integration, Ireland is one of the most open economies in the world. The Irish economy's success story in the 1990s and 2000s was strictly connected with its export performance and capacity to attract FDI. Ireland ranks fifth in the world for both exports and direct investments as share of GDP: total export reached 101% of GDP in 2010, with 55% accounted for by goods and the rest by services;40 inward FDI reached 121.3% in 2010.41 It also has the highest share of employment in foreign affiliates in the OECD (OECD, 2010). Indeed, over the last 20 years, Ireland has had remarkable success in attracting FDI in high-technology export-oriented sectors (particularly in pharmaceuticals, chemicals, computers, electronic machinery) and finance. Ireland is a preferred location for US multinationals' hubs in Europe, and US multinationals have typically been among the largest investors in the country (Economist Intelligence Unit, 2008); the UK and the Netherlands also played an important role. The direction of foreign trade also reflects the greater role played by the US than in other EMU countries, as well as the particularly strong connection with the UK.42 Moreover, the bulk of Irish exports are actually produced by foreign-owned firms - 90% in 2008, according to Brennan and Verma (2010), meaning that a much higher than usual share of Irish GDP accrues to foreign residents.43 The high dependence of Irish economic prosperity on international trade and investment flows helps to explain the priority it assigns to preserving credibility vis-a-vis external creditors, even at a high domestic cost, and in spite of the fact that most of its bonds were held abroad when the sovereign debt crisis started. Indeed, at the end of 2010, non-Irish residents held 82% of the long-term government bonds outstanding (Killian, Garvey and Shaw 2011). In other words, Ireland had too much at stake to risk jeopardising relationships with external financiers and trade partners by a sovereign debt restructuring or default. "Issue inter-linkages" and political bargaining gave external creditors an indirect voice in the political process. Indeed, as the next few pages will also show, Ireland's main trade partners are also its main creditor countries. Moreover, 40% of Irish corporate tax receipts are paid by US-owned companies (Economist Intelligence Unit, 2008), and generally the benefits from external interaction are diffused across the Irish population: 46% of manufacturing jobs and 27% of service sector jobs are in foreign affiliates (OECD, 2010). This ensures a fairly broad-based consensus across the economy in support of policies aimed at protecting external interests.44 It is telling that the Irish government maintained the 12.5% corporate tax rate, a key policy measure to attract and retain foreign business, even in the face of important fiscal tightening elsewhere in the economy, and vehemently defended it when European partners tried to impose an increase in corporate taxes as a condition for the EFSF loan.
These external considerations contributed to motivating Ireland's resulting endeavours to maintain credibility vis-a-vis external creditors and EU institutions during the debt crisis. This in turn led to additional public finance consolidation plans, to an attempt at transparency with regard to the magnitude of the banking sector issues and their likely impact on public finances, to an ongoing collaboration with the relevant EU institutions, and to frequent and relatively timely communication with the markets. This attitude appeared to help in reassuring markets during the first stages of the crisis, although it lost some traction once the bond sell-off accelerated. Meanwhile, because of its high level of integration, the size of the financial system, and the very large size of its gross external debt relative to GDP (1019% of GDP in 2010),45 Ireland was more important for the rest of the global economy than its relatively small GDP and GNP, as well as its population, would have suggested. For a start, the EMU and international banks were much more exposed to Ireland than to Greece by the time of the sovereign debt crisis. By September 2010, the size of foreign banks' exposure to Ireland was almost double that in the case of Greece, totalling 417 billion euros,46 in spite of Irish GDP being about two-thirds of Greek GDP. Among EMU banks, German banks were the most exposed to Irish debt in absolute terms (114 billion euros, equivalent to 4.7% of German GDP), adding an important incentive to the key source of funding for EMU bail-outs. The Netherlands, normally a crucial ally in Germany's austerity drive, also had a significant exposure in GDP-adjusted terms (14.8 billion euros, equivalent to 2.6% of Dutch GDP). France, in contrast, was the largest overall creditor of Greece. Moreover, claims from non-EMU countries were larger than EMU country claims in absolute terms. Not surprisingly given historical links, geographical proximity and the trade links shown above, the UK stood out as the largest overall creditor (118.5 billion euros, equivalent to 7.1% of UK GDP). The nationalised UK bank Royal Bank of Scotland (RBS) owned Ulster Bank, one of Ireland's "big four" banks, and was by far the most exposed foreign bank to Irish sovereign debt at the time of the European bank stress tests published in July 2010, with a reported total of approximately 5 billion euros (Murphy, 2010). This was potentially a major headache for the UK government.
The financial and trade exposures so far described clearly highlight how Ireland had a much greater importance than Greece for some of the major European and global economies. However, besides the specific case of RBS, the figures leave an impression of asymmetrically much greater importance of the rest of the world for the small and open Irish economy than vice versa. The Irish problem carried much greater indirect weight in European and global policymakers' considerations due to fears of international financial contagion. The Irish crisis carried significant systemic risk: a default in the intertwined system of Irish sovereign and bank debt had the potential to significantly damage an already impaired global financial system. Other vulnerable European banks would be affected, with the Spanish banking sector first in the firing line, creating an unstoppable chain of sovereign and bank defaults across the continent, and possibly on the other side of the Atlantic as well. In this light, it is not surprising that EU partners and the IMF did not need much convincing to provide rescue funds for Ireland. Additionally, Irish policymakers and EMU partners managed to collaborate quite swiftly, when compared with the Greek struggles. Subsequent announcements of fiscal adjustment paths by the Irish government ware generally endorsed by European authorities, such as, for example, the four-year plan presented in November (Ireland Department of Finance, 2010e). The Irish sovereign demonstrated capacity and willingness to deliver fiscal consolidation and generally stick to plans agreed with EMU partners, and this helped to ensure collaboration:47 it ensured EU and ECB endorsement for its adjustment plans and facilitated agreement on the conditionality of the rescue loan with an ambitious four-year plan.48 Press reports suggest that the ECB started considering using the EFSF to support Ireland as soon as the facility became operational in late September 2010 (Bloomberg News, September 27, 2010b). The dialectic of external partners trying to persuade Irish policymakers to accept external aid and Ireland reluctantly accepting is an interesting reflection of how intertwined the nature of interests on both sides had become, as well as of the power of persuasion held by external actors vis-a-vis the Irish government.
The higher level of proximity of Ireland to its creditors, as compared with the Greek situation, helps to explain the more collaborative attitude of both external and domestic actors during the sovereign debt crisis, which contributed to its relatively quicker resolution. Disentangling the direct effect of creditor/debtor proximity from other factors impacting Irish bond spreads during the sovereign debt crisis is not always possible. That said, bond markets appeared to recognise at least some of that international political economy dynamic, since they started to calm down as soon as evidence of an external push to access aid was accompanied by even vague Irish statements of ongoing collaboration, while they had treated similar statements issued during the Greek sovereign debt crisis with much greater scepticism. On top of EMU country interests, the sizeable exposure of UK banks and the interests of US multinationals provided strong incentives for the international community as a whole to favour external support for Ireland.49 The important role played by external economic and financial links in the bail-out decision is also shown by the UK contribution to the Irish rescue package,50 while only EMU countries had provided bilateral loans in Greece's rescue. Indeed, in the event, the 85 billion euro package included contributions from the UK, Sweden and Denmark as well as from the EFSF, the ESM and the IMF and from Ireland's own National Pension Reserve Fund and cash reserves.51 Of the total rescue package, 35 billion euros was earmarked for banking system support, and the rest would go to finance the Irish state.
Integrating the lessons from the Irish experience with a comparison with developments in Greece suggests that the higher level of economic, financial and ideological proximity of Ireland to its main creditors was positive for Irish sovereign credibility in bond markets during the sovereign debt crisis, other factors being equal, and as compared with the Greek experience. These findings, along with the results of the Greek case study, provide support for the prediction of hypothesis 3.2. In addition to the role of economic and financial inter-linkages, the EMU institutional backdrop was also more favourable to a quicker resolution at the time of the Irish crisis than when Greece was first hit. In fact, in the intervening period, some progress had been made to develop institutions and procedures to deal with crises at the EMU level. While Greece had needed to obtain official financing on the basis of protracted bilateral negotiations with EMU partners, a centralised fund, the EFSF, was now available to provide external assistance if needed, and the ECB was now open to buying peripheral bonds in secondary markets through its SMP when needed to stabilise markets. The activation of the EFSF only required Eurogroup approval (on top of IMF Board approval for the IMF part). This was much less problematic than ex-novo bilateral political negotiations, since the notion of intra-EMU rescue transfers was no longer under discussion, and conditions for access were now regulated by the EFSF statute. The centralisation of decisions and negotiations on the rescue package removed the need to embark on extensive bilateral negotiations when it came to accessing external finances, reducing the potentially disruptive role of creditor country domestic politics. Accordingly, markets did not have to deal with conflicting signals from the international arena, and rapidly accepted that a rescue plan was in the pipeline as soon as the ECB and EMU partners started discussing the possibility in early November 2010.