Ireland
Ex Post Evaluation of Exceptional Access Under the 2010 Extended Arrangement

This paper presents an Ex Post Evaluation of the 2010 Extended Fund Facility (EFF) arrangement with Ireland. The Fund approved in December 2010 an exceptional access EFF arrangement for SDR 19.466 billion (2,321.8 percent of quota) in support of Ireland’s home-grown program and as part of a broader financing package of Ireland and its European partners. The program focused on addressing the Irish banking crisis to break the adverse feedback loop between banks, the sovereign, and the real sector. It aimed to restore the banking system to health, including by establishing a smaller banking sector with high capital buffers and more stable funding sources; and to secure fiscal sustainability while limiting the near-term demand drag from fiscal consolidation. Large external financing was a key element of the crisis response. Program implementation was very strong. The program succeeded in stabilizing the banking sector and reducing its size, and fiscal developments were also broadly as anticipated. Domestic demand was, however, weaker than programmed and unemployment remained high, amid a very challenging external environment. Program success, including regaining market access at low interest rates, benefitted also from actions at the wider euro area level. The Ex Post Evaluation draws several lessons from Ireland’s experience under the EFF: ? The main lessons emerge from what worked well: Strong country ownership, setting (and meeting) realistic and tailored targets were key for success, combined with effective communication and pro-active engagement. Addressing a banking crisis requires strong and credible actions upfront. ? Some areas offer lessons for future program design: While the main pillars of the financial sector program were sound, more proactive and stronger supervisory interventions and other supportive steps could have strengthened banks’ balance sheets and bank profitability and helped resolve problem loans; bank recapitalization should be limited to those with viable medium-term business strategies; unsecured and non-guaranteed creditors of failed banks should be bailed in, provided a strategy to ring fence potential systemic risks can be put in place; macro-financial linkages require careful attention and timely steps to limit sovereignbanking sector feedback loops; fiscal policy has to be mindful of debt sustainability but also of domestic demand conditions, and it needs a clear anchor. ? There are also lessons related to Fund policies: Ireland’s EFF underscores the importance of addressing shortcomings of the systemic exemption clause in Criterion 2 of the exceptional access criteria; and it suggests the need to explore ways to secure stronger upfront commitments from monetary union authorities, when those are critical for program success.

Abstract

This paper presents an Ex Post Evaluation of the 2010 Extended Fund Facility (EFF) arrangement with Ireland. The Fund approved in December 2010 an exceptional access EFF arrangement for SDR 19.466 billion (2,321.8 percent of quota) in support of Ireland’s home-grown program and as part of a broader financing package of Ireland and its European partners. The program focused on addressing the Irish banking crisis to break the adverse feedback loop between banks, the sovereign, and the real sector. It aimed to restore the banking system to health, including by establishing a smaller banking sector with high capital buffers and more stable funding sources; and to secure fiscal sustainability while limiting the near-term demand drag from fiscal consolidation. Large external financing was a key element of the crisis response. Program implementation was very strong. The program succeeded in stabilizing the banking sector and reducing its size, and fiscal developments were also broadly as anticipated. Domestic demand was, however, weaker than programmed and unemployment remained high, amid a very challenging external environment. Program success, including regaining market access at low interest rates, benefitted also from actions at the wider euro area level. The Ex Post Evaluation draws several lessons from Ireland’s experience under the EFF: ? The main lessons emerge from what worked well: Strong country ownership, setting (and meeting) realistic and tailored targets were key for success, combined with effective communication and pro-active engagement. Addressing a banking crisis requires strong and credible actions upfront. ? Some areas offer lessons for future program design: While the main pillars of the financial sector program were sound, more proactive and stronger supervisory interventions and other supportive steps could have strengthened banks’ balance sheets and bank profitability and helped resolve problem loans; bank recapitalization should be limited to those with viable medium-term business strategies; unsecured and non-guaranteed creditors of failed banks should be bailed in, provided a strategy to ring fence potential systemic risks can be put in place; macro-financial linkages require careful attention and timely steps to limit sovereignbanking sector feedback loops; fiscal policy has to be mindful of debt sustainability but also of domestic demand conditions, and it needs a clear anchor. ? There are also lessons related to Fund policies: Ireland’s EFF underscores the importance of addressing shortcomings of the systemic exemption clause in Criterion 2 of the exceptional access criteria; and it suggests the need to explore ways to secure stronger upfront commitments from monetary union authorities, when those are critical for program success.

Introduction

1. This paper presents an Ex Post Evaluation of the 2010 Extended Fund Facility (EFF) arrangement with Ireland. In support of an Irish adjustment program and in parallel with Ireland’s European partners, the IMF provided an EFF arrangement to Ireland with exceptional access of SDR 19.466 billion (2,321.8 percent of quota), covering the period from December 2010 to December 2013. Under Fund policies, an Ex Post Evaluation of an exceptional access program reviews performance against program objectives, discusses if the program design was appropriate to address Ireland’s challenges, and if program modalities were consistent with Fund policies.

2. The paper is organized as follows. It begins with a brief discussion of developments preceding the Fund arrangement and initial domestic efforts to mitigate the crisis. This is followed by an overview of the key program objectives and financing parameters, and by a review of the main features of the program and of program outcomes. A subsequent section covers key program design issues and to what extent the program appropriately addressed Ireland’s challenges. A final section discusses the lessons that emerge from the Evaluation.

Pre-Preprogram Context: Unfolding of a Banking Crisis and Initial Efforts to Mitigate the Crisis

Following a history of strong growth and economic progress, imbalances began to build in Ireland after the turn of the millennium. The global financial crisis of 2008 exposed deep fault lines in the Irish banking system and triggered a severe banking and economic crisis.

A. The Irish Miracle

3. Ireland had a long history of exceptionally strong performance. Real GDP growth averaged around 5 percent and employment increased strongly for several decades prior to 2000, as the economy benefited from a successful integration into the world economy. Supported by high inflows of foreign direct investment (FDI) attracted by a competitive and business friendly environment, exports reached more than 100 percent of GDP and their structure changed towards high-value added products (IT, services, pharmaceuticals). Economic growth was accompanied by a marked decline in the public debt ratio, which fell in the three decades prior to the millennium from 110 (one of the highest among advanced countries at the time) to under 25 percent of GDP.

A01ufig01

Ireland and Selected Countries: GDP per Capita

(Ratio to US GDP per capita)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Sources: WB WDI database; and IMF staff calculations.
A01ufig02

Ireland’s Export ratio and Harmonized Competitiveness

(Percent of GDP)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Sources: Central Statistics Office of Ireland; and Central Bank of Ireland.

B. Domestic Demand Boom, 2000-07

4. In the wake of euro entry, Ireland’s economy underwent a process of change in the sources of growth. Following euro entry, Irish interest rates fell markedly and, with some lag, the economy experienced a major boom in domestic demand. This was supported by a large credit expansion and a property price driven wealth effect. Rising wages and prices depressed Ireland’s competitiveness (total factor productivity turned negative), leading the current account from balance in 2000 into deficits (-6 percent of GDP in 2008).

A01ufig03

Ireland: Growth Accounting, 1989-2007

(Percent)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Sources: KLEMS database; and IMF staff calculations.

5. The banking system played a key role in financing the boom and became very large by any standard. Credit to the private sector expanded by 20 percent per annum during 2002-07, and total assets of domestic banks reached over 520 percent of GDP by 2008. Weak regulation, supervisory oversight, and enforcement as well as growing competition contributed to a weakening of underwriting standards and lowering of interest margins, including on so-called “tracker” mortgages which carried narrow spreads over ECB rates.1 Commercial property loans to developers also expanded rapidly. The credit boom was financed through extensive recourse to cross-border wholesale markets, with wholesale funding reaching some 60 percent of banking liabilities.

A01ufig04

Total Assets of Domestic Credit Institutions and Private Sector Growth

(Billions of euro)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: Central Bank of Ireland’s Monetary Survey.
A01ufig05

Irish Banks’ Reliance on Wholesale and Eurosystem Funding

(Billions of euro)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: Central Bank of Ireland.

6. Emerging risks were, for the most part, not recognized at the time and not addressed. Ireland’s long history of economic success may have contributed to the fact that risks were largely not recognized or downplayed, both by domestic and foreign observers.2 This also applied in general to Fund surveillance prior to 2008, although staff reports noted that the “impressive” economic performance of Ireland was “increasingly reliant on house building.”3 The 2006 Financial Sector Stability Assessment, while highlighting vulnerabilities, did not raise significant flags.4

C. Crisis Triggers and Initial Containment Efforts

7. The global crisis had major repercussions on the Irish banking sector. The Irish banking system, outsized in proportion to the domestic economy and carrying large structural imbalances became vulnerable to a sharp correction in the property market and to disruptions in international capital markets. With high household indebtedness and leveraged real estate developers, the property price boom had already begun to slow in 2007, with residential and commercial property prices peaking that year. The widespread market tensions following the demise of Lehman in the fall of 2008 catalyzed a rapid worsening of the global market environment, in particular for banks such as Irish banks that were reliant on wholesale funding.

A01ufig06

Housing Over-valuation Started to Reverse in 2007

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Sources: PTSB/ESRI; CSO; and Haver Analytics.1/ Average house prices divided by moving 4-quarter adjusted GDI per capita.2/ Census 2011 rent extrapolated by CPI rental component over CSO house prices.

8. Banking sector difficulties and the bursting of the real estate bubble took their toll on the broader economy. Real GNP contracted by 9½ percent in 2008–10 (GDP by 9 percent),5 led by a 41 percent collapse in investment, in particular in construction. The economic downturn and falling property prices resulted in a surge of payment defaults, rising bank losses, and growing difficulties for banks to secure wholesale funding.

9. The Irish authorities gradually ratcheted up actions to restore confidence in the financial system. Major steps included (see Table 1): (i) a blanket guarantee scheme (CIFS) to depositors and creditors of six domestic banks in September 2008, which however intensified the nexus between the banking and sovereign balance sheets;6 (ii) nationalization of the failed Anglo Irish in January 2009; (iii) recapitalization of Bank of Ireland (BoI) and Allied Irish Bank (AIB)—the two largest domestic banks—by the state in February 2009; (iv) a new scheme, replacing the CIFS, in December 2009, covering newly issued and acquired bank liabilities; (v) the establishment of a state-owned restructuring agency, the National Asset Management Agency (NAMA), in December 2009; and (vi) recapitalization and nationalization of Irish Nationwide Building Society (INBS) in December 2009. In particular, NAMA purchased banks’ troubled large (in excess of € 20 million) commercial real estate (CRE) loans to developers at steep discount (relative to banks’ book value, although at a premium to market value), requiring in turn considerable state support for the banks. By mid-2010, the government had injected €46.3 billion of capital into banks (roughly 30 percent of GDP), de facto nationalizing domestic banks.

Table 1.

Key Milestones of the Irish Crisis, International and European Context

article image

10. As the banking crisis escalated, liquidity support provided by the Eurosystem increased sharply. With the continuing lack of market funding access and large outflows of wholesale deposits by corporates, banks increasingly relied on central bank funding to replace maturing liabilities. Irish domestic banking system exposure would peak at €90 billion through regular ECB operations, and at €150 billion (close to 100 percent of GDP) when adding the Emergency Liquidity Assistance (ELA) provided by the Central Bank of Ireland under its financial stability mandate. As a result, ECB exposure to Ireland was much larger (in relation to GDP) than for any euro area country through 2012.

A01ufig07

Eurosystem Financing of the Banking System in Selected Euro Area Countries

(Percent of GDP)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: IMF staff calculations.

11. On the fiscal front, the authorities implemented strong upfront measures to contain the deficit. The original 2008 budget had been expansionary, but amid collapsing revenues and the rising cost of supporting the financial sector, consolidation measures of 6.2 percent of GDP were implemented between 2008 and 2010. These included reductions in the public sector wage bill, cuts in capital expenditures, cuts and other restrictions on social welfare benefits, as well as revenue measures. In this period, expenditures in nominal terms shrank by over €5 billion. However, revenues continued to decline even faster, and the deficit (excluding financial sector support) reached 13.3 percent of GDP in 2010. Public debt jumped from 25 percent prior to the crisis to close to 100 percent of GDP in 2010. Overall, Ireland’s fiscal position turned out to be much weaker than suggested by earlier headline balances of close to zero or positive and low public debt (Box 1).

A01ufig08

Sources of Increase in Gross Debt to GDP Ratio

(Percent of GDP)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Sources: Department of Finance; and IMF staff estimates.

12. The initial crisis response proved insufficient to stabilize the economy, and the authorities requested a Fund-supported program in late 2010. Rising euro area market tensions, particularly in several southern euro member countries, the unraveling of the property market, uncertainties about the potential scale of bank losses, as well as contracting economic activity and rising unemployment all undermined the initial success of the domestic adjustment efforts. Sovereign bond yields came under pressure as the substantial public sector support to the banking sector and the collapse in tax revenues severely weakened Ireland’s fiscal position. With declining fiscal buffers and great concern expressed by the ECB7 about the scale of the liquidity support provided by the Eurosystem, the Irish authorities requested the support of the IMF and of their European partners. While Fund staff had intensified surveillance and cooperation with the authorities during their initial attempts to respond to the crisis, the request for a Fund program came only in late 2010, some three years after the onset of the crisis.

13. Some steps taken during this period would limit subsequent policy options for the Fund-supported program. These included the blanket guarantees extended to bank creditors, which provided some bank creditors time to exit and with banking sector burden sharing covering only a relatively modest amount of subordinated debt, these steps intensified the nexus between the banking and the public sector. Moreover, the state support for the banking sector contributed to a large increase in public debt to close to 100 percent of GDP (not counting the cost of further bank recapitalization), before the Fund-supported program went underway. As a result, the fiscal space for maneuver was severely limited under the Fund-supported program.

Program Strategy and Financing: Overview

The Fund approved an EFF arrangement in support of Ireland’s home-grown program in December 2010. Part of a broader financing package of Ireland and its European partners, the program focused on addressing the banking crisis in order to break the adverse feedback loop between banks, the sovereign, and the real sector. Fiscal targets were calibrated so as not to unduly undermine growth, while ensuring progress toward fiscal sustainability over the medium term. Program risks were regarded as high. A large amount of external financing was a key element of the crisis response.

A. Program Strategy

14. The program aimed to restore Ireland’s banking system to health and put its public finances on a sound footing, thus renewing confidence and a return to strong, sustained growth.8 The home-grown program built initially on the authorities’ multi-year National Recovery Plan (NRP) and, following the change in government in early 2011, on the new authorities’ Towards Recovery: Programme for a National Government and its Medium-Term Fiscal Statement 2011 (MTFS). It aimed to address head on Ireland’s key challenges:

  • Establish a smaller, robust banking sector focused on serving Irish economic needs. Banks would maintain high capital buffers and establish a more stable funding base.

  • Secure fiscal sustainability over the medium term, while aiming to limit in the near term the demand drag from consolidation, to the extent feasible. The authorities’ plans would form the basis of the 2011 and subsequent budgets.

  • Preserve a strong business environment. As Ireland was generally regarded as strong in this area, actions under the program were limited to addressing remaining impediments to competitiveness, employment, and growth.

15. Program conditionality was aimed at securing these objectives. Structural conditionality was focused on a comprehensive restructuring of the banking sector and steps to strengthen banking supervision (Tables 2 and 3 and Figures 1 and 3). In the fiscal area, two quantitative performance criteria were established, covering the central government cash primary balance and arrears; there was also an indicative target on central government net debt (Table 4). The use of the primary rather than an overall balance measure reflected large uncertainty about the path of interest rates. A symmetric adjuster for revenues (and a limited number of expenditure items) was designed to accommodate the budgetary impact of weaker or stronger economic activity. A small number of structural fiscal benchmarks and prior actions were also included, including submission of the 2011 budget.

Table 2.

Prior Actions under the 2010 EFF

article image
Source: IMF MONA database.
Table 3.

Structural Benchmarks under the 2010 EFF

article image
article image
Source: IMF MONA database.

Central Bank directions were issued within the required timeframe; however it was agreed to postpone completion of the capital injections required until after the General Election, to let the new government decide on the recapitalization. These directions were superseded by the Central Bank’s PCAR directions of March 31, 2011. The recapitalization of the banks took place in July 2011, following the PCAR.

Table 4.

Quantitative Performance Criteria and Indicative Targets under the 2010 EFF

(in billions of Euros) 1/

article image
Source: IMF MONA database.

Non-accumulation of new external payment arrears was a continuous PC, and was met.

Measured by the exchequer balance excluding interest payments. Cumulative from the start of the relevant calendar year. The target was adjusted for payments for bank restructuring and for revenue over- or under-performance (from March 2011 onward) according to the TMU.

Adjusted targets are shown in italics.

Figure 1.
Figure 1.

Main Features of Selected Recent European Programs

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Figure 2.
Figure 2.

Ireland EFF: Macroeconomic Outcomes, 2009–14

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Sources: CSOI; and IMF staff calculations with 2014 WEO data reported under ESA2010.
Figure 3.
Figure 3.

Financial Sector Conditionality Relative to Other Crisis Programs

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

16. Conditionality in the Ireland program was more streamlined than in other recent cases and tailored to address Ireland’s needs. More than in other recent European program cases, conditionality focused on financial sector reforms (see Figure 1), but the number of structural benchmarks was overall lower than in most recent SBA and EFF programs, including in other euro area countries (see text chart). The tailored use of conditionality reflected, consistent with the conditionality guidelines, Ireland’s relatively strong institutional framework (outside the financial sector). Moreover, Ireland’s flexible labor market and competitive business environment was judged as being in little need of reform. The authorities’ plans served effectively as the basis for the fiscal program and for other measures over the program period, helping to secure ownership by the authorities.

A01ufig09

Quantitative Performance Criteria in Recent and Past Programs

(Cumulative number of conditions divided by years covered by conditionality)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: Mona database, and IMF staff calculations. All GRA programs since 2002.
A01ufig10

Structural Conditionality in Recent and Past Programs

(Cumulative number of structural benchmarks divided by years of conditionality)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: Mona database, and IMF staff calculations. All GRA programs since 2002.

B. Financing, Exceptional Access, and Program Risks Financing

17. The Board approved a three-year exceptional access EFF arrangement for an amount of SDR 19.5 billion (€22.5 billion, or 2,321.8 percent of quota). The Fund’s EFF arrangement was part of a financing package of €85 billion; the EU and bilateral European lenders pledged a total of €45 billion, and the Irish authorities contributed €17.5 billion from the nation’s cash reserves and other liquid assets. Fund access relative to quota was the second largest of recent crisis cases (after Greece’s SBA in 2010) and the third largest relative to GDP. Upon approval, SDR 5.0 billion (399 percent of quota) was disbursed. At 26 percent of total access, the frontloading (measured by the first purchase) was around the median level of recent crisis cases, and the highest of euro area cases. Given the medium-term nature of Ireland’s balances of payments (BoP) need, a relatively long EFF program (of 36 months) rather than a 24-month SBA was considered appropriate (see below).

A01ufig11

Ireland EFF: Duration, Access, and Frontloading

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

A01ufig12

Total Access 1/

(Percent of Quota)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: IMF MONA database.1/ All GRA arrangements with exceptional access since end 2007excluding precautionary
A01ufig13

First Purchase Disbursement

(Percent of Access)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

18. The access request to an extended arrangement was justified by a large and medium-term BoP need and included a buffer to manage financial sector risks. Staff argued that Ireland’s BoP need reflected several factors:9 (i) a financial account deficit arising, in particular, from rollovers of external government debt and outflows from Irish banks; (ii) a current account deficit, which would only gradually decline; and (iii) the need for the authorities to strengthen their reserves in order to meet the government’s external debt obligations and shore up confidence. A notional buffer of €35 billion was available to support the banking system, with €25 billion expected to be needed for immediate recapitalization and the remaining €10 billion available on a contingency basis.

Exceptional Access

19. At the time of the EFF approval, staff argued—and the Board agreed—that all four criteria for exceptional access were met, reflecting the following considerations:10

  • Criterion 1—the member is experiencing or has the potential to experience exceptional balance of payments pressures on the current or capital accounts resulting in a need for Fund financing that cannot be met within the normal limits: the rationale set out in the staff report was summarized above (see paragraph 17).

  • Criterion 2—there is a high probability that the member’s public debt is sustainable in the medium term. However, in instances where there are significant uncertainties that make it difficult to state categorically that there is a high probability that the debt is sustainable over this period, exceptional access would be justified if there is a high risk of international systemic spillovers: Ireland’s exceptional access was justified on the basis of high risk of international systemic spillovers, as public debt was not judged to be sustainable in the medium term with high probability.

  • Criterion 3—the member has prospect of gaining or regaining access to private capital markets within the timeframe when Fund resources are outstanding: successful program implementation was expected to be the basis for future access to private capital markets on manageable terms.

  • Criterion 4—the policy program for the member country provides a reasonably strong prospect of success, including not only the member’s adjustment plans but also its institutional and political capacity to deliver that adjustment: staff judged that while the program risks were substantial, Ireland’s sound institutional and consensus-based approach provided reasonably strong prospects of success.

20. Issues related to Criteria 2-4 will be discussed in a subsequent section. In particular, these relate to whether the “systemic exemption clause” continued to be met at later program reviews (Criterion 2); and whether the program at its outset provided reasonably strong prospects of success in the absences of more upfront commitments to address tensions in the wider euro area (Criteria 3-4).

Risks

21. Program risks were recognized as high.11 Uncertainty about bank losses, a challenging public debt trajectory, downside risks to economic growth, and a pending general election were seen as the main domestic risk factors. The program provided a buffer for some of these risks, notably through additional capital for banks against a stress scenario. To mitigate election related political risks, the staff team sought ex ante support for key program parameters from all major political parties. Euro area wide risks did not feature prominently among the risks discussed in the staff report for the program request.12

Program Implementation and Outturn

Program implementation was very strong, and all performance criteria and most of the structural conditionality were met. The EFF arrangement was successful in stabilizing the financial sector, reducing its size, and containing systemic risks. Public debt and deficit developments were also broadly as anticipated. Low cost market access for the sovereign was regained by the end of the program period, benefitting also from actions at the wider euro area level. However, unemployment remains high and domestic demand was lower than programmed, amid an external environment that was much more challenging than originally projected. As was anticipated to some extent at the outset, substantial further efforts are needed in the post-program period to reduce significant remaining vulnerabilities, in particular in the fiscal area and the banking sector.

22. The track record of meeting program conditionality was very strong. All performance criteria under Ireland’s EFF were met at each of the reviews (Table 4). Of the structural conditionality, some 97 percent was met (Tables 2 and 3); albeit a few with some delay or partially; only one structural benchmark was not met.13 While a decision was taken to combine the first and second reviews in light of the calling of early elections, all other reviews were concluded as originally scheduled and the final and 12th review was concluded on December 13, 2013.

A01ufig14

Compliance with Quantitative Performance Criteria 1/

(Percent of total)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: IMF Mona database; and IMF staff calculations.1/ Weighted average of performance criteria, whereby a weight of 2 is assigned to fully met conditions, a weight of 1 to performance criteria met after modification, and a weight of zero to waived or not met performance criteria.
A01ufig15

Implementation of Structural Benchmarks1/

(Percent of total)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: IMF Mona database; and IMF staff calculations.1/ Weighted average of assessed benchmarks, whereby a weight of 2 is assigned to fully met benchmarks, a weight of 1 to benchmarks met with delay, or partially met, and a weight of zero to not met benchmarks.

A. Macroeconomic Framework and Structural Policies

Program Strategy

23. The macroeconomic strategy aimed at securing sustainable growth amid major headwinds facing the Irish economy. Fiscal adjustment was to proceed in broadly equal steps per annum, cushioning the pro-cyclical impact relative to a more front-loaded consolidation effort. As part of the banking reforms, steps were also taken to limiting adverse spillover effects on the domestic economy, including by ring-fencing the deleveraging to mainly non-core (i.e., foreign) banking assets. Nevertheless, the program anticipated a further decline in real GNP in 2011, bringing the estimated cumulative decline to almost 20 percent relative to the GNP level in 2007, to be followed by a gradual rebound in subsequent years.

24. Outside of the financial sector, the program included a limited set of structural reforms to strengthen the Irish economy. While the external current account deficit had reached 6 percent in 2008, competitiveness issues were generally not perceived as very significant. Ireland placed consistently among the top European countries in competitiveness rankings and among the most business friendly countries worldwide (see Figure below). Aside from steps to ensure better enforcement of competition regulations and some service sector reforms,14 structural reforms focused mainly on labor market policies. Initially, this included steps to reduce the public sector wage bill (but an agreed cut in the minimum wage was rejected by the new coalition government and reversed). Later into the program, high unemployment led, among others, to steps to strengthen active labor market policies and to reduce the minimum wage for new job entrants.

A01ufig16

World Bank’s ease of doing business ranking: Ireland vs. key European peers (2007-2014)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: World Bank.
A01ufig17

External Balance and Competitiveness

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Sources: CBI; CSO; Haver Analytics; and IMF staff calculations.1/ Four-quarter current account in percent of four-quarter GDP.

Outturn

25. The economy stabilized broadly as anticipated in the early program period, but weak domestic demand was a drag on the recovery in later years (Figure 2 and Table 5). The external sector provided a much stronger than expected impulse initially, as Ireland’s export exposure (40 percent to the US and the UK) and product specialization sheltered it from the slowdown in the euro area. Subsequently, however, external sector contributions to growth were lower than programmed following further trading partner weakness and the “patent cliff”15 that affected pharmaceutical exports strongly. Domestic demand was generally somewhat weaker than programmed. Overall, GDP growth was below the original program assumptions (0.9 percent per annum during 2011-13 versus 1.7 percent).16 This said, deviations from the original projections were comparatively limited in the case of the Ireland EFF, especially when considering the weaker external environment, and did not warrant revisions to the policy program. Ireland’s current account position improved substantially, notably due to a large increase in undistributed profits of redomiciled multinational firms,17 but as anticipated in the access discussion, the financial account recorded sizable net outflows.

Table 5.

Ireland Selected Economic Indicators, 2008-14, Outturn Versus Program Request Macro Framework

(Annual percentage change unless indicated otherwise, grey background for program request numbers, PR)

article image
Sources:Bloomberg; Central Bank of Ireland; Department of Finance; International Financial Statistics; IMF staff estimates.

Contribution to growth.

Excludes bank restructuring costs.

Data refers to end-December for nominal effective exchange rate, end-January for real effective exhcnage rate, end-February for private

Adjusted growth rate of credit to households and non-financial corporations.

General government balance per ESA2010 definition. For 2013, includes exchequer outlays for guarantees paid out under the ELG

A01ufig18

Weighted Real GDP Growth of Ireland’s Trading Partners

(WEO forecasts vs. outcomes) 1/

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

1/ Weights: Euro Area −0.5, UK −0.2, USA −0.3.
A01ufig19

Household Consumption After Cycle Turns

(Real Consumption: Peak Year 0=100)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Sources: OECD; national statistics; and IMF staff calculations.
A01ufig20

Gross Fixed Investment After Cycle Turns

(Real Investment: Peak Year 0 = 100)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

26. The slow recovery in domestic demand likely reflected more protracted repercussions from domestic balance sheet weaknesses than originally anticipated. Ireland’s domestic demand contraction lasted longer than in other advanced economies that faced a financial crisis and private debt overhang in recent years (see text figure). In these countries, private consumption growth typically turned positive again some three years after the peak. The Ireland EFF already anticipated a slower response (private consumption was to turn positive in 2012, i.e., five years after its peak). However, the simultaneous and very large deterioration of government, corporate, and banking balance sheets, combined with a very large household debt burden, were likely more detrimental to investment, consumption, and overall growth than originally anticipated.18 The intensifying euro area crisis likely also weighed on domestic confidence in the early years.

A01ufig21

Household Debt

(Percent of disposable income)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Sources: OECD; Haver Analytics; national statistics; and IMF staff calculations.

B. Financial Sector

Program Strategy

27. Through decisive, early actions, the financial sector strategy aimed to achieve a smaller, more robust banking system that would serve Irish economic needs. The program was heavily frontloaded, with most of the conditionality in the first year and reduced conditionality from the middle of the program onward. The strategy had three main objectives:

  • Separate non-viable banks from viable banks and return the latter to healthy entities.

  • Downsize the banking sector; and

  • Reduce systemic risk and increase the banking sector’s shock-absorbing capacity.

28. The core policy actions of the financial sector program were to:

  • Wind down Anglo-Irish bank and INBS, which were considered non-viable.

  • Subject the remaining intervened banks (AIB, EBS, BoI, and PTSB) to an early, stringent prudential capital assessment review (PCAR) to determine recapitalization needs.19 The PCAR included an independent diagnostic asset quality review and, building on this review, a stress test that covered a three year horizon. Capital needs were to be determined against both the base and stress case, and the results were to be disclosed within three months of the program.

  • Promptly increase capital buffers. Based on the outcome of the stress test, the recapitalization would be completed within the first six months. Banks were required to observe the Core Tier 1 (CT1) ratio of 10.5 percent in the base case on an ongoing basis and 6 percent in the stress case.

  • Subject the intervened banks to a prudential liquidity assessment review (PLAR) with the aim to reduce their reliance on short-term wholesale funding, including central bank funding, and ensure convergence with Basel III liquidity requirements over time. The CBI prepared the PLAR based on banks’ deleveraging plans and set liquidity targets for each bank. Banks’ deleveraging plans covered detailed strategies to downsize their balance sheet, including identifying noncore and core assets for disposal and run off or prepayments, with the aim to achieve a target loan-to-deposit ratio (LDR) of 122.5 percent by the end of 2013. 20 Banks were required to identify the liquidity and capital implications from deleveraging and factor these into their capital requirements.21 The LDR targets were dropped by the 7th review and replaced with monitoring of nominal targets for banks noncore asset sales.

  • Take steps to address distressed debts of households and SMEs. The program favored a combination of options aimed at a balanced restructuring process between creditors and borrowers while minimizing moral hazard. This included reforming the personal insolvency regime (structural benchmark), introducing a central credit registry (structural benchmark), and recapitalizing and strengthening the capacity of banks to undertake voluntary restructurings on a case by case basis. The Fund provided considerable technical assistance in this area.

  • Address institutional weaknesses. To make the financial system more robust, steps in this area were to include strengthening of banking supervision; establishing a comprehensive bank resolution framework, including a special resolution scheme for deposit taking institutions.

Program Outturn

29. Implementation of the financial sector program was largely as envisaged and banks capital buffers were increased significantly. The early stress test and the independent diagnostic asset quality review provided a credible basis for recapitalizing domestic banks. The three intervened banks were recapitalized within six months of the onset of the program; capital buffers roughly doubled and have remained well above regulatory requirements.22 With BoI able to attract some private capital and the success of liability management exercises, the capital injections by the public sector were below initial program assumptions. The banks that were assessed as non-viable, Anglo-Irish bank and INBS, were merged and subsequently liquidated, and NAMA’s asset disposals proceeded somewhat faster than initially envisaged. Finally, steps to strengthen financial supervision were also implemented as programmed.

30. Banks improved their funding structures and restored access to capital markets. Mainly through balance sheet deleveraging and some deposit mobilization, banks’ loan-to-deposit ratio declined significantly, from 190 percent at end 2010 to under 120 percent at the end of the program in late 2013. Consistent with program objectives, banks reduced their reliance on short-term wholesale funding, and their recourse to the Eurosystem declined markedly, from a peak of €90 billion to €26 billion at end-2013.

BOI, AIB, and PTSB Aggregated Financial Statements, 2010-2013

article image
Sources: CBI; and IMF staff estimates.

31. Banks sharply downsized their balance sheets, mainly through the disposal of non core foreign assets. The scaling back of the outsized Irish banking sector was the most aggressive among the stressed EA countries. Total assets were reduced by 20 percent,23 with the three intervened banks selling or disposing € 45 billion in assets. Most of the targeted non-core foreign assets were ultimately sold at higher prices than expected in the restructuring plans. At the same time, banks also shrunk their domestic loan book between 2010 and 2013 by more than 25 percent, a somewhat higher reduction than projected in banks’ original deleveraging plans. The LDR targets may have contributed to an excessively rapid deleveraging in banks’ loan book and were dropped by the mid-2012). Banking assets at end 2013 stood at about 2.5 times GDP, less than half their ratio prior to the program.

A01ufig22

Changes in Irish and EU Bank Exposures, 2010Q4-2013Q2 1/

(Percent)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Sources: European Banking Authority; and GFSR (April 2014).1/ Based on a sample of large banks in each region (including two Irish banks).

32. Notwithstanding these major achievements, banks still face significant challenges. At the outset of the program, it was decided not to address some policy challenges upfront but to leave those for a later date. This decision reflected a mixture of capacity and political constraints, as the early steps to stabilize the banking system required major technical, legal, and political efforts. It was also recognized that no quick fix solutions existed in some areas. Reflecting these deliberate early decisions and an inability to tackle and resolve these issues later on, the program ended with the banking sector continuing to face a number of major challenges, including:

  • Little progress was made in addressing and resolving rising SME and mortgage arrears (Figure 4). Amid a weak economy and rising unemployment, NPLs continued to rise until the end of the program period, reaching 27 percent of total loans at end-2013 compared to 10 percent at end-2010. Households and SMEs accounted for 60 percent of the end-2013 NPL stock. Other factors such as strategic defaults may also have contributed to the rise. Generally, banks were slow in providing restructuring solutions to their borrowers, with over 70 percent of the end 2013 NPL stock 720 days or older. The incipient recovery and more prescriptive stance of the CBI, including by introducing quantitative SME and mortgage arrears restructuring targets and further strengthening of loan provisioning and classification standards, helped advance loan restructuring starting mid-2013.24 About one-quarter of mortgage arrears were restructured by end-2013, and repeat defaults have remained low.

  • The implementation of a new personal insolvency framework and the central credit registry did not progress as originally envisaged. A new personal insolvency regime went into effect only in 2013. While the staff team had favored a court-administered restructuring framework that would provide incentives for effective out-of-court debt settlement and enforcement arrangements, the authorities ultimately did not adopt a court-based framework The reform shortened, however, the discharge period from 12 to 3 years (after which the liabilities under the insolvency proceedings would be discharged) and introduced the option for out-of-court debt and enforcement procedures for restructuring mortgages, subject to creditor approval by certain majorities. The credit registry, an important mechanism for creditors to assess borrower’s consolidated financial situation, was not implemented by program end. Slow progress in these areas may have weakened incentives for banks to engage borrowers on debt restructuring.

  • While improving, bank profitability was not restored by the end of the program period. Banking sector losses declined and intermediation margins improved over the program period. However, high NPLs and associated provisioning costs limited progress in this regard, and low returns on sizable asset positions (including on tracker mortgages) combined with limited progress in improving efficiency and reducing operating cost continue to weigh on banking profitability. As a result, banks reported losses for a sixth consecutive year in 2013. This contributed to a gradual decline in their CT1 ratio from 16.5 percent at end-2011 to 13.4 percent at end-2013, although these ratios remain well above the regulatory minimum requirement. Forward looking profitability concerns are most pronounced at the smallest of the three domestic banks (PTSB).

Figure 4.
Figure 4.

Developments in Arrears and Terms of Restructuring

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: Central Bank of Ireland; and IMF staff calculations.
A01ufig23

Return on Equity

(After provisioning, in percent)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: Bankscope.

C. Fiscal Policy

33. Building on the authorities’ plans, the fiscal program aimed to address fiscal sustainability concerns while mitigating adverse effects on growth. The authorities’ plans set out a three-year adjustment objective of €15 billion (9 percent of [2010] GDP) in measures relative to a baseline scenario, with €6 billion to become effective in 2011. This would steadily reduce the overall deficit while stabilizing the debt ratio by 2014 at somewhat above 120 percent of GDP. Expenditure reductions would account for two thirds of the adjustment, covering capital spending, public sector staffing and salaries, as well as savings on social welfare and public pensions for new entrants. Revenue measures primarily aimed at broadening the income tax base and increasing its progressivity. From 2013 and beyond the program period, further revenue measures were to raise indirect taxes through phased increases in VAT rates and the introduction of a new property tax.

34. The pace of fiscal adjustment and the appropriate fiscal anchor were subject to considerable debate with the authorities and among the Troika partners. In designing the fiscal program, Fund staff focused principally on the size of adjustment measures relative to the authorities’ baseline as set out in their NRP. Troika partners, however, focused mainly on the fiscal deficit, and in particular on the timeframe needed to reach the 3 percent of GDP deficit threshold of the Stability and Growth Pact (SGP). After considerable debate and amid concerns about the weak growth outlook, it was agreed to delay the attainment of the 3 percent SGP threshold to 2015, one year later than originally targeted in the NRP.

35. A limited set of structural reforms focused on further strengthening fiscal institutions and frameworks. These included a new Fiscal Council, a fiscal responsibility law, and a medium-term fiscal framework, all structural benchmarks and aimed at strengthening fiscal credibility. Revenue administration and expenditure management were considered generally strong and were not subject to program conditionality. Privatization targets were not central to the fiscal program, given the relatively limited stock of non-financial sector state owned assets.

Outturn

36. Fiscal indicators improved significantly and generally in line with the original program targets. All quantitative fiscal targets were met. Between 2010 and 2013, the overall deficit narrowed by 3.4 percentage points of GDP, and the primary deficit by 5 percentage points. Broadly in line with expectations, the gross debt ratio increased markedly to 124 percent of GDP. As noted earlier, Troika partners focused on different fiscal targets; but with fiscal developments broadly on track, the risk that this could result in conflicting messages did ultimately not materialize.25

37. Lower-than-anticipated interest payments did not lead to a corresponding improvement in the overall balance, and the primary deficit was larger than originally envisaged. Considerable interest savings relative to the original program path (a cumulative €4.4 billion during 2011-13, or 2.7 percent of 2013 GDP) reflected in part lower-than-programmed interest rates as well as the re-profiling of the European partner support. The interest savings were largely offset by weaker revenues, which underperformed also in relation to GDP. As a result, lower interest payments did not lead to a stronger overall fiscal balance, which would have limited adjustment needs for future years. The improvement in the structural primary balance was still some 1½ percent of GDP per annum during the program period, but this pace was some 1 percentage point per annum slower than in the two years prior to the EFF-supported program, when the authorities had attempted to contain the crisis without external support.

Interest Payments: Original Program vs. 12th Review

(Euro billion, unless otherwise indicated)

article image

38. Structural fiscal reforms were implemented as originally programmed. To improve expenditure management, three-year aggregate and ministerial level expenditure ceilings were established and put on a statutory basis. In 2012, a Fiscal Responsibility Act was passed and included, among others, a general government budget balance rule and debt rule consistent with the SGP. The Fiscal Responsibility Act also established the independence and legal basis of the Irish Fiscal Advisory Council (IFAC). The IFAC is responsible for providing an ex ante assessment of the macroeconomic forecasts underpinning the budget and for assessing the soundness of the government’s budgetary projections and fiscal stance. Measures were also adopted to enhance transparency, including improved fiscal reporting, forecasting, and risk management.

Program Design Issues

This section discusses if the program design was appropriate to address Ireland’s challenges and if program modalities were consistent with Fund policies. The focus will be selective and not cover a number of areas where program design issues would not seem to arise. Moreover, the issue of why certain program commitments were not observed—a key feature in some earlier Ex Post Evaluations of Fund programs—is also not covered: it is largely moot in light of Ireland’s strong record vis-à-vis the program conditionality (a later section will include lessons that can also be learned from Ireland’s robust implementation record).

A. Financial Sector Strategy

Should more have been done to address distressed household and small enterprise debt?

39. The early recapitalization of banks did not accelerate an orderly restructuring or resolution of distressed debt. The program provided banks upfront with a fairly robust capital cushion, based on results of the initial stress tests and asset quality review. Nevertheless, and notwithstanding some supervisory guidance, banks generally did not employ the capital buffers to proactively restructure distressed household debt, and significant reforms to facilitate debt restructuring were put in place only in 2013. The slow progress reflected in part legal issues, including delays in introducing the personal insolvency regime and the central credit registry. 26 Historical and political constraints made some options seemingly not viable in the Irish context, including repossessions of a significant number of owner-occupied residences. Moreover, the introduction of the one year moratorium on foreclosure of residential mortgage arrears as well as a high court ruling in 2011 which made it more difficult for banks to initiate summary foreclosure proceedings on mortgages taken out before 2009 may have led some borrowers with adequate financial resources to default on their loans (“strategic defaulters”).

40. Progress in this area was slow, but the time was used to build a broader consensus on a restructuring reform strategy. As noted, political and legal constraints contributed to slow progress in this area, raising in part moral hazard concerns, yet there was also little if any fiscal space available for generalized debt relief to the private sector, as had been done in some countries (Box 3). In the end, the program period and lessons learned from initial attempts were used to build a fairly broad consensus around a menu of options, which may offer better prospects for addressing these issues going forward. In many respects, Ireland broke new ground in this regard in the euro area—but to get to this point, it took some five years since the property price bubble began to burst.

41. While no “quick fix” solution existed and distressed debt resolutions take time, there are some areas where faster progress could have contributed to a stronger economic recovery. Mindful of the case for a consensus-based approach, there are still areas where more decisive steps would likely have made a material impact on macroeconomic performance. Over time, this could have helped restore banks’ role in credit intermediation sooner and mitigated the negative effect of household and SME debt on economic activity. Such earlier steps, which were extensively debated during later program reviews, could have included:

  • More forceful supervisory intervention and prudential guidance. In particular, this could have (i) addressed deficiencies in banks’ provisioning treatment of impaired assets, including restructured loans; (ii) provided guidance to banks on loan restructuring options, including any debt discharge, to liquidity constrained but ultimately viable borrowers; (iii) set sooner realistic targets for the reduction of problem loans; and (iv) encouraged write offs of uncollectible loans.

  • Putting in place a personal insolvency regime that balances creditor and debtor interests. A strong personal insolvency regime can provide a clear framework for resolving distressed personal debt and set out an appropriate balance between creditor and debtor’s interests. Legal amendments in this area were delayed, and the ultimately adopted framework, while breaking new ground in the euro area, did not incorporate some relatively fast and objective resolution mechanisms (see also ¶31), which a few other countries have found useful.27 The amendment to the Land and Conveyancing Act in 2013 strengthened, however, the options to pursue foreclosures as part of the resolution process, which could be particularly useful in the case of loan defaults on investment (“buy-to-let”) properties.

Should the program have included a greater focus on improving bank profitability and identifying and resolving non-viable banks?

42. The weak profitability of the banking system remains a concern. Under the Fund-supported program, rapid progress was made in addressing immediate systemic stability issues in the banking sector, as discussed above. However, all intervened banks continued to record a loss (net of provisioning) throughout the program period, as operational and funding costs weighed on banking profits, combined with large holdings of low yielding assets, especially on tracker loans and non-income earning NPLs. Banks also sold relatively profitable and well performing foreign assets as part of their deleveraging plans, with the proceeds mainly used to reduce comparatively low-cost ECB funding. The loan-to-deposit ratio targets, in place for part of the program period, also constrained banks domestic activities and led to upward pressures on deposit rates in 2011–12 as competition for banking deposits increased. Along with the deep economic contraction, all these factors tended to adversely impact bank profitability. While the early focus on immediate stability concerns was appropriate, a stronger effort could have been made, especially once these concerns had been addressed, to achieving sustained improvements in bank profitability. Aside from addressing NPLs and other distressed debt, discussed previously, this could also have included stronger attention to better aligning banks’ operating cost structures (e.g., compensation policies and branch networks) with their lower revenue streams. More substantial progress on containing banks’ operational costs began in 2013.

43. A related issue concerns the need to identify early on whether a bank is viable or not, and limit recapitalization with public funds to viable institutions. All domestic banks required considerable public sector support. While a decision was made to wind down Anglo Irish and INBS, the program decided to keep other banks as going concerns and recapitalize these banks with public funds. The recapitalizations were based on a careful asset quality review and stress tests, but these by their very nature did not address the issue whether the banks had a viable business model going forward. Indeed, the medium-term viability of one of the three remaining banks (PTSB) still remained to be established when the program ended.28 In the meantime, PTSB was recapitalized in July 2011 by the government with €2.7 billion, net of receipts of €1.3 billion from the sale of its life insurance business, but recorded net cumulative losses of €1.7 billion during 2011-13.

44. The failure to restore banks to viability or, alternatively, to wind all non-viable institutions down early, contrasts with best bank resolution practices.29 The lack of progress in restoring PTSB to viability or in winding it down reflected divergent views on how best to proceed, and on how to secure the required financing. Considerations favoring to keep PTSB open included the EC’s Directorate-General for Competition’s view that a third domestic bank was essential for competition.30 As concerns financing, it proved difficult to secure the upfront funding for either a break-up of PTSB into a good bank and a bad bank, as the government had proposed, or for winding down PTSB. If no external funding could be secured, there was also the view that pushing the resolution decision to the future may be less costly to the taxpayer than an upfront wind down, given highly distressed market conditions and the relatively low net present value of PTSB’s tracker loans. Despite extensive discussions, these issues could not be resolved among the Troika partners and the authorities, and PTSB continued to have recourse to ECB bank funding facilities. The Fund-supported program ended with the acknowledgement that PTSB faced the greatest challenge in restoring profitability and was not expected to break even after provisioning expenses until 2017.31

Did the de-leveraging of foreign assets contribute to cross-border financial fragmentation?

45. The banking sector strategy called for the rapid sale of mostly noncore foreign assets. The motivation for this strategy was threefold. First, with assets of around 520 percent of GDP, domestic banks were considered too large from a financial stability perspective. Second, the sale of assets was the only viable option to better align the size of assets with the banks’ stable funding sources, and reduce reliance on wholesale funding and temporary ECB liquidity support. Third, with domestic markets depressed, the sale of Irish assets, including of mortgages, was not an option to advance a rapid asset deleveraging.

46. The sale of foreign assets left Irish banks more exposed to domestic risks. While the deleveraging strengthened the banks’ liquidity position and reduced size-related systemic risk this also led to a greater concentration of banks’ risk exposure to the domestic market. Moreover, the sale of foreign (“non-core”) assets by Irish banks as well as by banks in some other euro area member countries tended to add to fragmentation in the region. A more domestically focused banking system potentially aggravates sovereign-banking feedback loops.

47. On balance, however, the deleveraging strategy was reasonable—but safeguards would be required if it were applied to a larger economy. Notwithstanding some drawbacks for the profitability and risk concentration of Irish banks, the overall emphasis on deleveraging an outsized banking sector was sound. The safeguards against fire sales provided also important protection against downside risks. It must be recognized, however, that the strategy did not internalize spillover effects on other countries. While this did not have major adverse effects given the size and exposure distribution of Irish banks, spillover issues would need to be addressed if such strategies were applied to several countries or to one of the larger economies.

Was the bail-in of private creditors too limited?

48. Senior creditors did not contribute to the banking sector resolution in Ireland. During 2008-11, subordinated debt in Ireland was subject to sizable so-called “liability management exercises” that generated a reduction in bank liabilities of about €15.5 billion (some 10 percent of GDP).32 However, senior creditors did not take part in these exercises and, following the issuance of a blanket guarantee in 2008, a sizable portion of senior bondholders redeemed their positions by the time of the Fund-supported program in late 2010. Nevertheless, an intense debate at the time concerned the remaining bondholder exposures (totaling some €16.4 billion), and if these bondholders should be bailed in as part of the banking sector resolution strategy.

49. There were strong arguments in favor of bailing in the senior unsecured bondholders in the intervened banks. In an idiosyncratic bank failure, uninsured creditors are generally expected to absorb losses.33 This reduces the risk of moral hazard and should help contain the risk of future crisis. It also contains the costs borne by the public sector and would thus weaken the banking-public sector links that otherwise can undermine public debt sustainability (as was the case in Ireland). And while the exposure of senior unsecured bondholders was limited by the time the Fund-supported program got underway, it was still equivalent to more than the full size of the fiscal measures during the program period. Finally, the bail-in of senior unsecured bond holders could be seen as part of an equitable burden sharing process.

50. Concern about cross border euro area spillovers was the main argument against a bail-in. The debate about bail-in of senior unsecured bondholders took place at a time of major banking fragilities in the euro area. There were concerns that a bail-in of senior bondholders could be seen as precedent setting in the euro area and adversely affect euro area banks and their funding markets given also extensive cross holdings by banks of other euro bank senior bonds. The lack of fully-built up euro area firewalls and of a clearly articulated euro area policy for addressing spillover risks contributed to contagion concerns in late 2010-11. Aside from concerns about cross-border spillovers, the bail-in of senior unsecured bondholders can also raise concerns if a failed bank is of systemic importance domestically. However, this was not a major issue in the Irish case as the state intervened in all domestic banks and most domestic risks were thus already addressed.

51. The evidence is not clear, however, on the risks of cross border spillovers from bailing-in senior unsecured creditors in Ireland, and policies could have been put in place to address these risks more directly if they arose. Spillovers should have been limited if markets and bondholders of Irish senior unsecured bank debt were expecting a bail in. Indeed, Irish (senior unsecured) bank bonds traded at the time at levels consistent with clear anticipations of a principal haircut, reflecting that some burden sharing was anticipated by bondholders and markets. While the anticipation of risks does not always preclude additional repercussions if these risks actually materialize, the magnitude of the repercussions should generally be more contained. Moreover, even if cross border contagion risks were considered important, steps could have been taken to ring fence these through appropriate policy responses in the affected markets. This could have included supporting steps by country authorities in cases where their banks’ solvency would be threatened from writing down their direct exposures to Irish senior unsecured debt; and/or, if needed, by forceful liquidity support by the ECB to ensure no disruptions in euro banks’ funding markets.,

A01ufig24

Irish Banks Bonds Prices (senior unsecured debt), 2009-12

(Euro)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: Bloomberg.

52. The lack of senior unsecured bondholder bail-in continues to have adverse legacy effects. Aside from the direct implications for fiscal adjustment needs, the legacy effects relate principally to issues of perceived fairness and burden sharing. In particular, even if concerns about spillovers were paramount—a case that is, as discussed, not obvious for bailing-in unsecured creditors in a failed bank—many in Ireland question why Irish taxpayers should be the ones covering the cost of addressing such euro area wide concerns. A bail-in or other solution that would have “mutualized” these costs would likely have resulted in more equitable burden sharing. Moral hazard concerns were also an important argument for a bail-in of creditors at the time. However, such concerns were ultimately reduced with the new EU Bank Recovery and Resolution Directive, which includes senior bondholder bail–in in banking rescue operations, effective January 1, 2016.

B. Other Program Design Issues

Was the speed of fiscal adjustment appropriate?

53. Substantial fiscal adjustment needs were left for the post-program period. Gross public debt increased by almost 100 percentage points since 2007, exceeding 120 percent of GDP in 2013. In contrast to other euro area program countries, the primary balance remained in deficit and was not expected to record a surplus until the first post-program year. Moreover, the staff’s debt sustainability analysis at the end of the program period indicated that Ireland continued to face high risks to debt sustainability.34 As discussed above, these outcomes were broadly in line with the original program assumptions.

54. In light of the remaining significant vulnerabilities, a question arises if the fiscal adjustment pace should have been accelerated. There are strong arguments on both sides of this issue, which was intensely debated inside and outside the Fund:

  • Arguments for faster adjustment included that it would help contain vulnerabilities, including those that could arise from unforeseen events in a highly uncertain environment; that adjustment fatigue would inevitably make fiscal cuts harder in later years, suggesting a more front-loaded strategy; that the fiscal multiplier was relatively low in Ireland, given its open economy, implying lower adjustment costs than in many other euro area countries; and that there was room left in Ireland to cut spending (which had increased sharply in real terms in the prior boom years) or raise revenues (which remained low relative to GDP).

  • The arguments against more front loaded adjustment rested mainly on its adverse impact on an already very weak economy; on staying with the home-grown Irish adjustment program, thus benefiting from strong ownership; and on Ireland’s track record of delivering on planned adjustments, which was seen as providing some assurance that future adjustment needs would also be addressed, even if they arose outside the framework of a Fund supported program.

55. While a case can be made for moderately stronger fiscal adjustment once the economy stabilized, the broad balance of the program strategy appears sound. This view takes into account that Ireland’s fiscal adjustment was in any case sizable—some 1½ percent of GDP per annum in terms of the structural primary balance during the three program years, and a cumulative 10 percentage points during 2009-13. Especially in the first year of the program, when GNP had been expected to decline for a fourth year in a row (and by 18.7 percentage points relative to its peak in 2007), any additional fiscal drag could have proven self defeating. Moreover, the adjustment that was secured during the program, with two-thirds of the measures on the spending side, could be expected to have a durable effect. The case for moderately faster adjustment becomes, however, stronger toward the end of the program period, once tail risks (also in the euro area) had receded and the fiscal multiplier would likely have been smaller.35 For example, saving just over half of the interest savings vis-à-vis the original program assumptions in the final program year would have secured an improvement in the structural deficit of about 1 percent of GDP in 2013. It would have moved Ireland somewhat further along toward debt sustainability, and could have alleviated some of the concerns about adjustment fatigue that seem to have gained momentum in the post-program domestic debate.

Did the program have a sufficiently clear fiscal anchor?

56. The authorities and troika partners focused on a variety of different fiscal indicators. The authorities were guided by the fiscal objectives outlined in the National Recovery Plan and MTFS, which envisaged a €15 billion nominal adjustment from a baseline over four years. The plan also outlined medium-term objectives for the overall deficit and debt-to-GDP ratio. The European Commission focused principally on the overall deficit and the adjustment path under the excessive deficit procedure. As agreed, the program aimed to bring the deficit to no more than 3 percent of GDP by 2015. IMF staff focused on the nominal adjustment of the NRP (€15 billion over four years, with annual targets set for each year). The IMF program also set a performance criterion for the cumulative exchequer primary balance (with some adjustors, as noted earlier), and an indicative ceiling on the stock of central government net debt.

57. More clarity about the fiscal anchor could have been helpful. It is best practice to establish a clear anchor for fiscal policy.36 While the staff’s focus on the adjustment effort had some merit—e.g., it effectively allowed for the full play of automatic stabilizers—this indicator became increasingly difficult to quantify. It was measured relative to a hypothetical “no policy change” scenario, but such a scenario is affected by a variety of factors over time. Other measures, such as the overall (or primary) deficit, possibly allowing for automatic stabilizers, may have provided a clearer signal and measure of success. In any case, the use of different fiscal target measures by different partners and the authorities did not help in the communication of the fiscal strategy. Such differences could potentially have resulted in varying policy assessments, had the economy evolved very differently from its programmed path—a risk that ultimately did not materialize (except for relatively marginal adjustments; see ¶35), as fiscal developments unfolded broadly along the lines of the original program.

C. Was the Program Consistent with Fund Policies?

Were the Exceptional Access Criteria observed?

58. At the time of the EFF approval in December 2010, all four criteria for exceptional access were deemed to be met. The assessment for Criterion 1 was relatively straightforward as Ireland, notwithstanding its membership in a monetary union, experienced exceptional balance of payments pressures. The staff and Board also considered that the other criteria were met:

  • Criterion 2—while it could not be asserted that public debt was sustainable with high probability, exceptional access was justified by a high risk of international systemic spillovers. The spillover risks were mainly seen with respect to other euro area peripheral economies and, compared with other program requests resorting to the systemic exemption clause, the staff report for Ireland’s program request contained a relatively substantive discussion of these risks.37

  • Criterion 3—the program envisaged that progress on banking reform and the exceptional financing provided by the Fund and European partners under the program would drastically reduce debt default risks and normalize access to private capital markets within the timeframe when Fund resources are outstanding.

  • Criterion 4—the policy program was judged as providing reasonably strong prospects of success. While program risks were substantial, Ireland’s sound institutional and consensus-based approach was seen as providing a basis for a positive assessment with respect to this criterion.

59. Important shortcomings of the systemic exemption clause (Criterion 2) have been discussed separately by the IMF Board, and the staff reports in Ireland’s case did not present a strong case that the clause was met in later reviews. While discussions on possible reforms of the exceptional access framework are ongoing (see paragraph 64), the current policy requires that the systemic exemption must be met at every review to justify continued Fund financing in exceptional access cases, unless the country’s public debt is judged sustainable in the medium term with high probability. As European firewalls were established, and as the ECB expanded its toolkit and clarified its support for euro area members, tail risks in the area receded and the periphery began to stabilize. Under these circumstances, it is not obvious that spillover risks remained significant and that the systemic exemption clause was met at the time of the reviews in late 2012 and 2013 (see also text figures below which generally point to reduced spillover risks emanating from Irish banks and from foreign bank exposures in Ireland). The Board’s assessment in this regard could have been facilitated by an update of the staff’s analytical work for the original program request.

A01ufig25

Conditional Probability that European Bank(s) Fall(s) into Distress given an Irish Bank Falling into Distress

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: IMF staff calculations.
A01ufig26

Foreign Bank Exposures to Ireland 1/

(Billions of USD)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: Bank for International Settlements.1/ Ultimate risk basis.

60. With program success depending critically on actions by third parties that were not directly bound by the program, an issue arises if Criteria 3 and 4 were met. The Board noted that provision by the ECB of liquidity support was essential.38 But the issue went well beyond liquidity support for Ireland itself: it is difficult to see that the program, even if fully implemented by the Irish authorities, could have succeeded without further actions to stabilize the euro area and address tail risks. Such risks became increasingly apparent in the first half of the program period, as the euro area crisis intensified. In particular, absent euro area wide actions, there was little prospect of stabilizing Irish banks and establishing a sustainable funding base (essential for meeting Criterion 4), and of securing public debt sustainability and regaining market access at sustainable rates within the time frame when Fund resources are outstanding (required under Criterion 3). For example, in mid-2012 Irish bond spreads, while coming off their highs in early 2011, remained still about 200 basis points above the levels required to secure debt sustainability.39 The prerequisite actions were ultimately taken by the European partners,40 aimed at preserving the integrity of the euro area, and resulted in a re-tightening of spreads—not just for Ireland, but for the euro periphery more generally (see chart). Most studies suggest that such common factors (and not only country-specific actions) played an important role in reducing spreads in the euro area, possibly by more than 200 basis points for Irish yields (see text charts, and the text table for a sample of studies). Given the importance of euro area wide actions, the case for meeting the conditions of Criteria 3 and 4 would have been strengthened if the program had included upfront commitments to broader European and euro area actions (see also below), in particular once the euro crisis intensified.

A01ufig27

Five-year Government Bond Spreads to German Bund

(Basis points)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Sources: Bloomberg; and IMF staff calculations.
A01ufig28

Irish Bond Yields and Decomposition of Spread vis-à-vis Germany

(Cumulative change since 11/25/2011, basis points)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: Sgherri, Silvia, 2012, “Commonalities, Mispricings, and Spillovers: Another Look at Euro Area Sovereign Risk,” in 2012 Spillover Report-Background Papers.

Summary of Studies Assessing Impact of ECB Policies on Selective Euro Area Countries’ Sovereign Bond Yields or Spreads

article image
References:Altavilla, C., D. Giannone and M. Lenze, 2014 “The Financial and Marcoeconomic Effects of the OMT Annoucements,” Center for Studies in Economics and Finance, University of Naples, WP No. 352Purdue, W. and R. White, 2014, “What drove Irish Government bond yields during the crisis?,” National TreasuryKrishnamurthy A., S. Nagel and A. Vissing-Jorgensen, 2013, “ECB Policies involving Government Bond Purchases: Impact and Channels,” Working Paper, University of California, Berkley.Hochstein, M. 2013, “Fundamental Drivers of Euro sovereign risk permia,” Allianz Global Investors

Was the EFF the appropriate Fund instrument?

61. Ireland’s program broke new ground and became the first exceptional access EFF. In 2009, the Executive Board had expressed the expectation that the EFF would not normally be used to provide exceptional access financing.41 Indeed, the Board had approved in May 2010 for Greece a SBA arrangement, even though the focus on structural reforms—one key feature of the EFF relative to the SBA—was probably more central for the Greek program than it would be for Ireland.

A01ufig29

Ireland EFF: Principal Repayment to the IMF and the EC—SBA and EFF Terms, 2013 – 22

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Source: IMF staff calculations.

62. The use of the EFF represented an early lesson from the crisis—one that benefited Ireland first, but was subsequently also applied to Greece. The Board supported staff’s proposal for an EFF, which was principally anchored on a medium-term balance of payments need, as required under earlier Board decisions.42 The longer repayment period provided for under the EFF was also a contributing factor in securing favorable repayment terms from Ireland’s European partners, helping to address market concerns about recovery given the combined deleveraging of the public and private sectors. However, staff’s arguments in favor of the EFF versus an SBA (a longer repurchase period helps in resolving a medium-term BoP financing need; and the need for deep structural reforms) would arguably have applied even more strongly in the May 2010 program with Greece, and possibly some earlier Fund arrangements. All in all, the use of the EFF presented therefore in part an early lesson learned from the previous crisis programs—and issues of uniformity of treatment at least vis-à-vis the Greek case were ultimately addressed by the move to an extended arrangement for Greece as well.

Lessons

The lessons that emerge from Ireland’s engagement with the Fund under the extended arrangement are drawn to a good extent from what worked well under the program and the factors that contributed to Ireland’s very strong performance vis-à-vis program objectives. It should be recognized, however, that not all of these lessons are readily transferable to other Fund program cases and, as always, country specific circumstances need to be considered carefully. In particular, Ireland’s relatively strong capacity and (in most areas) robust institutions are likely to make some lessons less applicable to the typical SBA/EFF arrangement, where constraints in these areas tend to be more prevalent.

A. Lessons from What Worked Well

63. The very strong track record of implementing the Fund-supported program offers important lessons. Many of the “lessons” that emerge from the positive outcomes of the program have been well-established in earlier reviews of Fund-supported programs.43 The lessons include:

  • 1) Country ownership is key. The EFF arrangement supported for the most part a home grown policy program; in Ireland, implementation was already underway before the Fund-supported program commenced.

  • 2) Set realistic targets and meet them. While always an important objective, meeting program targets is particularly relevant for re-assuring private capital markets. Consistently implementing its commitments allowed Ireland early on to separate itself from other program countries in the euro area. Establishing a realistic but conservative macroeconomic framework was also critical in this regard.

  • 3) In a banking crisis, take strong actions upfront—credible asset quality and liquidity assessments and a well-capitalized banking system are critical. The asset quality and liquidity reviews, designed by the authorities in close cooperation with the Troika team and with experienced private sector involvement, established early on credibility with market participants. This was further strengthened when the capital increases, which were based on these reviews, provided some cushion against unforeseen events—even if these actions were costly in terms of public sector funds.

  • 4) Focus conditionality on key challenges. The parsimonious design of conditionality focused on the main challenges facing the Irish economy (financial sector and selected fiscal issues). It eschewed wading into deeper structural reforms and over-burdening implementation capacity, a judgment that also recognized the underlying structural strength and competitiveness of the Irish economy.

  • 5) Communicate effectively. While technical expertise is vital for the right diagnosis and in identifying the appropriate policy response, communicating the strategy is also of critical importance. The Irish authorities were very effective in communicating their program to the general public, market participants and vis-à-vis the Troika partners. On the Fund side, the staff team (including the Resident Representative) spent considerable time and effort on outreach activities, and this helped in achieving strong buy-in for the program strategy and the Fund’s role.

  • 6) Be pro-active and closely engaged. While earlier Fund surveillance largely failed to identify the building crisis risks, a staff team engaged pro-actively with the authorities by the time the crisis broke and the authorities began to respond, initially outside a Fund program. The closer interaction with the authorities during this time (and later with the Troika partners) laid the seeds of trust and mutual understanding that was critical once the program got underway.

B. Improving Program Design—Possible Lessons

64. The following lessons relate to program design and are based on a weighing of the pros and cons of alternative strategies. It speaks to the strength of the staff team and its Troika partners that each of the issues covered below was clearly recognized and extensively debated, including with the authorities. These lessons do not take issue with the main pillars of the authorities’ program—these were clearly sound—but cover important specific elements that may also arise in other Fund-supported programs.

  • 1) Bank recapitalization alone does not advance debt workouts and restore bank profitability—supervisory interventions and other supportive steps are also needed. The program rightly focused initially on addressing the immediate threats to the banking system, recapitalizing banks based on credible asset quality reviews. Subsequently, stronger supervisory guidance—for example, by fully using the room provided under IFRS and prudential regulations (as discussed in Box 2)—as well as further legal reforms could have facilitated more progress in addressing the high levels of NPLs in a sustainable manner. Earlier adoption of a stronger insolvency framework could also have facilitated progress in this area. Similarly, stronger supervisory guidance on bank restructuring could have helped in reducing operating costs and improving bank profitability. Taken together, more forceful actions, taken at an earlier stage, could have supported a stronger economic recovery.

  • 2) Bank recapitalization should be limited to viable institutions with a clearly sustainable medium-term business strategy. Recapitalizing a non-viable bank and/or not restructuring a bank that does not have reasonable prospects of being profitable over the medium term risks adverse repercussions for the entire sector and the economy at large. In the case of Ireland, a resolution of these issues should have been agreed during the program period.

  • 3) Unsecured and non-guaranteed creditors of failed institutions should be bailed in, accompanied by a strategy to ring fence potential risks when needed. Market participants generally anticipated a bail-in of unsecured creditors in failed Irish financial institutions. Moral hazard considerations, the need to contain public debt, and experience outside the euro area also argued for this approach, at least in failed institutions that required public support, and that it should have covered not only subordinated debt holders but also senior bond holders. While this approach would have broken new ground in the euro area at a time of area-wide fragility, negative spillover effects to other parts of the area, to the extent that they might have occurred despite the anticipation by market participants, should have been ring fenced at the euro area level.

  • 4) Macro-financial linkages require careful analysis, and timely steps should be taken to limit the sovereign and banking sector feedback loops. It proved difficult to identify the build-up of risks ahead of the bursting of the property price bubble in Ireland. This experience adds to the case for strengthening the analysis of macro-financial linkages, especially in countries with large financial sectors, and the toolkit to address related risks—an area where considerable efforts have been underway since the global crisis. Steps to sever or at least limit the sovereign-banking sector feedback loops need to be an important part of the toolkit.

  • 5) Fiscal adjustment has to be mindful of debt sustainability but, to the extent that this provides room, also of domestic demand conditions. On this basis, the original fiscal program design struck a broadly appropriate balance. The economy was extremely weak and tail risks were large in the early parts of the program. This said, and given sizable adjustment needs beyond the program period, a case could be made for a moderate tightening of the fiscal stance, especially in the final program year, once the domestic economy stabilized and European policy actions had taken major tail risks off the table. At that point, the fiscal multiplier in Ireland’s very open economy would likely have been small (but, of course, still positive). Saving at least half of the interest savings relative to the original program assumptions would have been one option and brought the primary balance closer to its original program target.

  • 6) Fiscal policy needs a clear and clearly communicated anchor. While all program partners agreed on the broad strategy, each adopted their own anchor and indicators to monitor performance. A unified approach would have helped communicate the program objectives more effectively and avoid possibly uncertainties and mixed signals, if program performance had deviated from the original targets (an issue that ultimately did not arise in Ireland).

C. Lessons Related to Fund Policies

65. The experience under the EFF arrangement offers also lessons related to Fund policies.

  • 1) The systemic exemption clause of exceptional access Criterion 2 has significant shortcomings. From the outset, staff considered that Ireland’s public debt could not be judged as sustainable with high probability in the medium term. Invocation of the systemic exemption at the original program request—given the euro area-wide risks then prevailing—meant that this vulnerability was not addressed at the time. The experience under Ireland’s EFF points to significant shortcomings of the systemic exemption clause, and discussions on possible reforms are ongoing at the Fund.44 Pending such reforms, the systemic exemption clause must be met at every review to justify continued Fund financing in these exceptional access cases, unless the country’s public debt is judged sustainable in the medium term with high probability. Consequently, when spillover risks appear to have diminished later in the program period, continued financial support by the Fund needed to be contingent on securing medium-term public debt sustainability with high probability. In the case of Ireland’s EFF arrangement, the application of the criterion was ultimately not covered in much detail in later reviews when spillover risks appear to have diminished.

  • 2) For a member of a monetary union, Fund-supported programs require close and effective interaction with the relevant union authorities. On Ireland, the Troika teams invested considerable time and effort into establishing a close and generally very effective working relationship. This relationship offered important benefits, including bringing in very large financial resources, both of the ECB and the EU, as well as to some extent complementary expertise of these institutions. Ireland’s policymaking is also strongly embedded in the European framework, and having its partners involved in the process brought to the table relevant EU-wide issues and constraints. This said, the experience under the Fund-supported program with Ireland highlights also several challenges, and the Fund’s objectives (focused on supporting its member country) may not necessarily always be aligned with those of the EC and ECB (generally focused more on EU and euro area wide concerns).45 An example here may be the bail in of senior unsecured bondholders, where European institutions focused on wider area concerns even if this resulted in a higher public debt burden for Ireland. The experience under the extended arrangement with Ireland suggests two main lessons:

    • Ways should be explored to secure stronger upfront commitments and understandings from monetary union authorities, where such steps are critical for program success. While sufficient steps were eventually taken, stronger upfront commitments in addition to financing assurances would have strengthened program planning and implementation. This could have involved, for example, broad policy understandings, at least among senior officials, of what it would take and what steps would be envisaged at the monetary union level to support the member’s program success. Such understandings could also have dispelled any doubts that the policy program provided reasonably strong prospects of success and of regaining market access (i.e., that Criteria 3 and 4 of the Fund’s exceptional access framework were met). Fund program documents should also cover in sufficient depth the risks for program success that emanate from a country’s membership in a monetary union.

    • Close and early Fund engagement with monetary union authorities can facilitate program success. There were little Ireland-specific interactions of the Fund with other Troika partners until late 2010, some two years after the property bubble began to burst. In the interim, each party was involved relatively independently with the Irish authorities. Earlier, close cooperation on emerging crisis risks, as well as on other key issues both at the national and union-wide level, could have facilitated faster progress in developing policy solutions and possibly stronger program results.

Estimating Ireland’s Structural Fiscal Balance

Pre-crisis estimates of the Irish structural fiscal balance did not capture the extent of the deteriorating fiscal position. In particular, estimates for the structural balance did not adequately highlight Ireland’s growing dependence on cyclical and asset price related tax revenues.

A01ufig30

Structural primary balance

(Percent of Potential GDP)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

Once the crisis had taken hold, estimates for Ireland’s structural balance were revised often and by large magnitudes. For example, the 2007 Article IV staff report estimated the structural balance in the previous year to be a surplus of 1.7 percent of potential GDP. During the 2009 Article IV, this balance was revised downwards to a deficit of 6.6 percent. Subsequently, this estimate was considered to be too pessimistic and was revised upwards. Similarly, the 2008 structural primary deficit was in 2009 estimated to be almost 14 percent of GDP. It was revised often during the program, with the final estimate settling at 10.5 percent of GDP (see first text chart).

Difficulties with estimating Ireland’s output gap posed major challenges for identifying the underlying structural fiscal position. Prior to the crisis, staff believed that both actual and potential output were close to each other. This assumption was based on the observation that Ireland’s labor market was closely integrated with that of the UK and open to labor inflows from new EU member states. These factors were seen as ensuring that unemployment was close to its natural rate and that the output gap tended to be small.

A01ufig31

Output Gap

(Percent of Potential GDP)

Citation: IMF Staff Country Reports 2015, 020; 10.5089/9781498359733.002.A001

The pre-crisis estimation method regarded changes in asset prices as structural in nature and the vulnerability of revenues to changes in asset prices was under-estimated. In the decade prior to the crisis, residential investment and house prices soared. As a result, property-sensitive taxes—stamp duty and capital taxes, and VAT— increased faster than GDP.

These deficiencies in the estimation method were exposed as the crisis unfolded. The magnitude of both the GDP decline and the asset price correction meant that staff struggled to find robust estimates of both the output and asset price gaps. This was one factor why the program was cast in terms of the annual quantum of fiscal consolidation measures, rather than in structural balance terms.

Could Irish Banks have Made More Progress in Addressing Problem Loans?

By the end of the Fund supported program, Irish banks had made little progress in addressing their problem loans. Despite relative robust regulatory capital levels, banks did not restructure a substantial amount of their problem loans (17 percent of mortgage NPLs at end- 2013 were restructured). Banks also did little in terms of writing off problem loans and collecting assets of long-term defaulted borrowers.

To an extent, these problems reflected weaknesses in the accounting treatment of problem loans and loss recognition under IFRS, which Irish banks follow, and in the EU’s loss impairment framework under the Capital Requirements Regulation and Directive. Under IFRS, banks are required to follow an incurred loss accounting model, which requires occurrence of a loss event, and financial assets are assessed for the existence of objective evidence of impairment. This can lead to “too little, too late” provisioning, as it does not account for estimates of future credit losses unless a loss event has already occurred, and provides considerable discretion in its implementation. The revised IFRS 9, which will come into effect in 2018, addresses some of the weaknesses by moving to an expected loss impairment model. However, there will remain a substantial role for supervisors as IFRS 9 continues to allow considerable discretion in bank management’s judgment.

Notwithstanding weaknesses in the current IFRS treatment of impaired financial assets, bank supervisors have the room to guide banks in their loan loss provisioning practices, given the imprecise nature of loan loss assessments and the lack of incentives by banks for self action. As highlighted in a recent IMF Working Paper, supervisors should, among others, make sure that banks’ overall provisions are sufficient and timely, and if warranted they can bridge any outstanding gaps between existing accounting regimes and prudential requirement.1 This is to ensure that banks do not overstate income and capital. Setting the right provisioning incentives helps accelerate the clean-up of banks’ balance sheets and restores also over time their ability to lend to the real economy. These recommendations are consistent with the 2002 issued BCBS Supervisory Guidelines for Identifying and Dealing with Weak Banks (currently under revision) and the Basel framework, where home supervisors are responsible for assessing credit risk and enforcing capital adequacy. As such, supervisors have considerable prudential powers in guiding banks to address asset quality problems. The BCBS supervisory guidelines recommend that weak banks with asset quality problems devise a remedial action plan that includes( i) negotiate new agreements with its viable but weak debtors; (ii) set time-bound targets on the reduction of problem loans; (iii) take possession of loan collateral and other debtor assets, when warranted; (iv) write off long-term problem loans; (v) use realistic loan loss provisioning against problem loans; (vi) classify restructured debt as NPL until the debtor demonstrates full capacity to repay at market rates.

Other factors may have also delayed the restructuring of distressed debt. These include banks’ limited capacity in restructuring large scale numbers of troubled borrowers; the delayed introduction of the credit registry which the program viewed as an important mechanism to enhance credit standards as it gives banks a consolidated picture of a borrower’s obligations across lenders; and the implementation of the personal insolvency took longer than expected. Other legal obstacles included the one year foreclosure moratorium on borrower’s that defaulted in CBI’s Code of Conduct for Mortgage Arrears and a high court judge’s decision in 2011 that did not allow for summary repossession proceedings for mortgages that originated before 2009. Recognizing the overall slow progress, the CBI pushed forward with a new arrears resolution strategy in 2013 (citing its supervisory powers provided by Regulation 70 of Licensing and Supervision of Credit Institutions) by setting performance targets for banks to reach on restructuring of SME and residential mortgage arrears in a sustainable manner and by tightening further its classification and provisioning standards on impaired assets.

In hindsight, Irish regulators had the room to act more forcefully and earlier during the program period, for example by using their prudential powers to strengthen classification of NPLs and restructured loans, provisioning and write off practices, and by setting restructuring targets.2 This could have encouraged banks to more expeditiously dispose of their NPLs or the underlying collateral. Allowing a significant amount of nonperforming assets to remain off the market also impedes asset repricing and investment in property markets. Faster progress could help restore debtor’s payment capacity while also allowing banks over time rebuild profitability. In turn, this could have underpinned the recovery in the broader economy.

1 See Supervisory Roles in Loan Loss Provisioning in Countries Implementing IFRS (WP/14/170).2 Supervisors must be willing to take timely and effective action; see IMF Staff Position Note, The Making of Good Supervision: Learning to Say “No”, May 18, 2010.

Selected Country Experiences in Restructuring Household Debt Overhang

In the wake of a banking crisis and sharp downturn in the economy and correction in housing prices, many countries in Europe as well as the United States faced widespread household financial distress. In some countries this led to a significant rise in mortgage arrears. In response, several countries put in place measures to address the household debt overhang. These mainly aimed to mitigate the negative macroeconomic and bank balance sheet effects from excessive household deleveraging and wide spread foreclosures.1 Large scale household debt distress was for many countries a novel feature and had also not been central in most earlier crisis episodes outside Europe, including in the Asian financial crisis and the Latin American debt crises.

Generally, there is no international best practice to household insolvency but rather a combination of measures is typically advocated to address household distress. Experience shows that countries choose a menu of policy tools to address household distress, and that the choice varies depending on a debtor’s and a country’s circumstances, including legal traditions and cultural preferences.2 It will also depend on the systemic scale of the problem, whether it involves household mortgages, the fiscal space, the legal framework, institutional capacity and arrangements, and the financial strength of banks or other household lending institutions.

Many countries have introduced or amended the personal insolvency regimes as a means to address in an orderly and equitable manner the restructuring of defaulted residential mortgages.3 A key issue facing most countries is that restructuring distressed mortgages has typically not been part of insolvency regimes. Reforming the regime to include mortgages has often met with considerable opposition by banks. There however have been precedents. For example, the insolvency framework in Norway which was introduced in 1993 following a banking crisis and housing market bust allows for both a voluntary and a compulsive court administered debt settlement. Under the court administered settlement, adjustments to mortgage debt are made to reduce the loan to value (LTV) to a maximum LTV ratio of 110 percent. Following a successful completion of a 5 year payment program (pay interest only on the secured part), the unsecured debt portion would be written off (discharged) if the borrower continued to own the house. Denmark has had a similar system in place since the mid 1980s, while Finland and Sweden adopted similar systems in the mid 1990s as well as more recently Iceland. The Irish authorities were more reluctant to move towards a court based resolution for mortgage debt, citing strong protections of private property rights in the Constitution and moral hazard risks as it could be costly to banks given that any households with negative equity could still service their debt and would gain unneeded relief or have strong incentives to default. Most of the reform efforts generally have focused around: (i) reducing the discharge period (i.e., the time it takes to emerge from the insolvency proceeding, whereby some or all of the debtor’s liabilities are written off) to give debtors a fresh economic start; (ii) defining eligibility requirements; (iii) defining capacity to repay; and (iv) introducing and defining automatic stay periods. In addition, a number of countries (including Ireland) have adopted out of court workout mechanism by issuing guidelines or establishing legally binding frameworks. Such a mechanism operates in the backdrop of an effective personal bankruptcy regime.

Although rare, some governments such as Iceland, Latvia, the United Kingdom and the United States have provided financial support to households to address the debt overhang. This ranged from establishing government funded asset management corporations that would purchase distressed household mortgages from banks and restructure the debt to restore the borrower’s payment capacity; to other measures where households were given permanent or temporary payment subsidies to help lower the present value of their loan (Table).

Finally, almost all crisis countries have also strengthened banks’ capacity to bilaterally and voluntarily restructure distressed residential mortgage debt. This includes recapitalization of banks, putting in place appropriate provisioning practices for impaired loans, strengthening regulatory guidance governing NPL resolution, and requiring the establishment of specialized loan workout units.

Policy Measures to Address Household Financial Distress and Encourage Restructuring of Household Debt Overhang

article image

On condition that payment capacity is restored. No DTSI limit specified.

1 See IMF April 2012, Dealing with Household Debt and Principles of Household Debt Restructuring SPN/09/15.2 For instance, in Europe most mortgages are recourse loans, meaning that after default borrowers are responsible for the difference between the value of the debt outstanding and value of the house. By contrast, in most states in the U.S mortgages are non recourse, in practice, and the difference is typically discharged. Non recourse loans can help lower the household debt burden faster, but at the same time can lead to higher NPLs and bank losses (see IMF Global Financial Stability Report, April 2011).3 See Y. Liu and C. Rosenberg “Dealing with Private Debt Distress in the Wake of the European Financial Crisis,” WP/13/44, IMF and K. Fletcher, “Housing Recoveries—Cluster Report on Denmark, Ireland, Kingdom of Netherlands—Netherlands and Spain,” 2014 Draft IMF Working PaperSource: IMF April 2012 Dealing with Household Debt; Principles of Household Debt Restructuring; CBI 2013 Mortgage Arrears Resolution Targets

Appendix. The Irish Authorities’ Views on the Ex Post Evaluation1

The Irish authorities broadly agree with the main findings of the Ex-Post Evaluation (EPE) report. We welcome this report and note that the focus is to assist the Fund in learning from the experience of the Irish programme of external support and to enable positive lessons to be implemented in future programmes. We appreciate the IMF’s recognition that the programme implementation was strong, and particularly agree with main lesson learned that strong country ownership, setting and then meeting realistic and tailored targets were key for success. We would nevertheless like to comment on a number of issues related to our programme, and the commentary on these which is included in the report.

Ireland’s Programme – Overview

Ireland’s programme had four broad aims - to address financial sector weaknesses, to raise Ireland’s growth potential, to strengthen our public finances, and to fully regain international capital market access. The programme is recognised to have been a success, with programme commitments delivered within the agreed deadlines. The year since exit has shown that the programme objectives have been met. The financial sector has been significantly restructured with considerably enhanced supervision. The recently completed ECB Comprehensive Assessment confirmed the capital adequacy of the two main banks. Order has been restored to the public finances, a primary surplus has been achieved, the debt has stabilised and is now firmly on a downward trajectory and Ireland is on target to bring the general government deficit below the Excessive Deficit Procedure (EDP) ceiling in 2015 as required. The economy is growing strongly again – by around 4.7% this year and is forecast to grow by close to 4% in 2015. Employment is growing strongly and unemployment is falling – although it remains at a high level. The success of Ireland’s return to the financial markets is characterised by the proposed early repayment of around €18.3 billion of our IMF loan of €22.5 billion – with €9 billion to be repaid in December 2014.

Timeframe of the Review

The Irish authorities understand that the timeframe of the review stops at end 2013, and is backward looking from that point. While this may be standard IMF methodology, it appears unusual that developments since that time are not taken into account, particularly as the report will be published more than one year after Ireland’s programme exit. Some of the value of the lessons learned identified from this approach is thus restricted and diminished.

The assessment of policy choices should have regard to the options available to Ireland at the time

The report includes an extensive discussion of the appropriate approach to bail in for Ireland. However, at the time of the banking crisis the option now included in EU law as part of the Banking Union at a European level - bailing-in the senior bondholders - was not available to the Irish Authorities despite the previous Government seeking this option for banks in wind down. The firm view at the time from the Troika was that sovereign support for banks was necessary to avoid contagion. This meant that the burden had to be borne by the equity holders, the junior bondholders, and particularly by the Sovereign which ultimately meant the Irish taxpayer. This approach has been highly controversial and is an issue which has yet to be fully resolved.

For Ireland, the EU approach is of particular relevance – and the measures now adopted under the Banking Union, including bail in provisions, if they had been available at the time, would have significantly reduced our overall debt. In the event, these policy options were not available, and the report’s references in this area must be viewed in terms of lessons learned for future programmes.

It is important to prioritise legislative commitments

Ireland’s programme was characterized by a very heavy legislative burden, reflecting, inter alia, the need for significant reforms in relation to financial sector regulation. The Report suggests that more could have been done earlier, particularly on the financial services side of the Irish programme. Ireland is acknowledged as having a well-functioning administration, but it must nevertheless be recognized that there are capacity limits to the amount of legislation that can be undertaken successfully, particularly when it is concentrated in a small number of sectors. Future programmes should have regard to such capacity limits both within administration and in the wider economy, and plan the phasing of legislative requirements accordingly, taking account also of national priorities in areas not covered by a programme.

The programme should be flexible and reflect each State’s individual market conditions and cultural sensitivities

EU-IMF Programmes are complex and involve the co-operation and agreement of many stakeholders. A degree of flexibility is required to ensure its policies are able to adjust to the multi-institutional complex circumstances faced during a programme. Although core objectives such as growth and debt sustainability are central, each country is unique and a one size fits all approach should not be utilized in programme design. Individual market and cultural sensitivities should be recognised.

The Report’s discussion of one domestic bank (PTSB) is outdated and should not be misunderstood as reopening settled matters

PTSB will continue to be an important participant in the residential mortgage market and a contributor to improving the level of competition in this market, which contracted very significantly in recent years.

During the EU-IMF programme the authorities worked on a wide range of proposals regarding the banking system generally and PTSB in particular. The strategy for PTSB agreed during the Troika programme was implemented diligently during 2012-2013 and has resulted in significant progress on many fronts, including a reduction in ECB funding of more than 70% from peak. It is accepted that further work is required by PTSB to return to sustainable profitability, a process which has progressed ahead of expectations to date.

While the ECB’s Comprehensive Assessment found PTSB to have sufficient capital in the base case, the adverse case calculation led to a capital increase requirement. PTSB has submitted a capital plan for ECB approval and should raise the capital within the timeframe required. We expect PTSB to play a meaningful part in the lending market, a key element in sustaining our economic recovery.

The focus on a proposed faster pace of consolidation is misplaced

It is important to note that Ireland’s external assistance programme started in the third year of a domestic fiscal consolidation process that had started in 2008. The final year of the programme represented the sixth year of consolidation in Ireland. Nevertheless, the crisis that faced Ireland in 2010 cannot be understated. Ireland was cut off from accessing funding on international markets, there was a growing cash shortfall in public finances, the economy was in deep recession and the banking system was in crisis. It was necessary to prioritise measures and to implement them quickly. The changes implemented were very significant and far reaching throughout society yet social consensus was maintained. Suggestions about a faster pace or greater scale of consolidation must be assessed in this context.

An appropriate balance was required to repair the fiscal accounts and yet at the same time support the aggregate demand. Frontloading was necessary to signal the authorities’ determination to put the public finances on a sustainable path. Additional frontloading would have been damaging as there was a need to maintain social cohesion and protect the fragile economic recovery.

Voted spending has been reduced by over €10bn from the 2009 peak of over €63bn to the present level of €53 billion. This reduction has been achieved while responding to the increasing need for public services and support in particular across areas such as welfare, health and education. Expenditure reductions were made on the basis of ensuring the fiscal target of deficit reduction was achieved each year while seeking to ensure that expenditure was focused on areas that protected the vulnerable, fostered growth and employment and facilitated reform of the public services. The report does not acknowledge the huge burden that was stoically borne by the Irish public. Any further frontloading of consolidation could have jeopardized social consensus and by extension the ultimate success of the program.

Ireland met and frequently outperformed its programme fiscal targets, both those set by the IMF through the quarterly Quantitative Performance Criteria and the annual deficit ceilings under the EU’s EDP. This strong fiscal performance was achieved despite the fact that growth was lower than projected when the programme was being agreed. In that context, the suggestion about the use of interest savings for debt reduction appears to ignore the lower growth outcome.

Ireland’s fiscal performance reflects the fact that the targets were realistic. The frequent outperformance, which effectively resulted in lower than expected debt levels, appears to be interpreted as an indicator that targets were insufficiently ambitious. The correct interpretation is that the targets were realistic and their achievement contributed to the positive perception of Ireland’s programme. In that context, the commentary in the report in relation to the application of reduced interest costs to debt reduction must be viewed as misplaced and redundant. The reductions in the interest costs for our EU loans were achieved following delicate negotiations with our European partners which were, incidentally, not reflected in similar reductions for the IMF loans.

Conclusion

We would like to thank the IMF staff for its professional and courteous behavior during the operation of the programme and for the objective policy advice that we received. The role of the Fund as a trusted advisor is highly valued by the Irish authorities.

1

More than half of mortgages contracted during 2005-07 were offered at 100-125 basis points margin over the ECB refinance rate (so-called “trackers”). Trackers accounted in 2011 for almost half of the mortgages for owner occupied houses and almost two thirds for buy-to-let properties; see IMF Country Report No. 12/147.

2

This was noted, for example, in P. Honohan, The Irish Banking Crisis Regulatory and Financial Stability Policy 2003-2008, May 2010; and K. Regling and M. Watson, A Preliminary Report on the Sources of Ireland’s Banking Crisis, May 2010.

3

See Ireland: Staff Report for the 2007 Article IV Consultation, IMF Country Report No. 07/325. Fund surveillance in the late 1990s/early 2000s had, however, flagged emerging risks, with for example the executive summary of the 2000 Article IV report noting that “… rapid growth in private sector lending pointed to a need for extreme vigilance on the part of market participants and supervisors alike to ensure that financial system soundness is not compromised”; Ireland: Staff Report for the 2000 Article IV Consultation, IMF Country Report No. 00/97.

4

While noting risks to financial stability “… there are several macro-risks and challenges facing the authorities… household debt to GDP ratios have continued to rise, raising some concerns about credit risks…a significant slowdown in economic growth, while seen as highly unlikely in the near term, would have adverse consequences for banks’ non-performing loans…” the report did not identify their magnitude and systemic nature:”…Stress tests confirm, however, that the major financial institutions have adequate capital buffers to cover a range of shocks.” Furthermore, the report was sanguine about the quality of bank supervision: “…good progress has been achieved in strengthening the regulatory and supervisory framework, in line with the recommendations of the 2000 FSAP.”See Ireland: Financial System Stability Assessment Update, IMF Country Report No. 06/292.

5

Gross Domestic Product (GDP) and Gross National Product (GNP) are two common measures of the economy and differ in the treatment of net factor incomes. GNP is the total income remaining with domestic residents and differs from GDP by the net amount of incomes accruing locally to non-residents or received from abroad by domestic entities operating overseas (net factor income). Mainly due to the role of large multinational companies operating in Ireland and related net factor income flows, GNP in Ireland is less than GDP (84 percent of GDP in nominal terms in 2013). As it captures the net factor income of domestic residents, GNP may be a better measure of the domestic activity and demand conditions in Ireland.

6

The Credit Institutions Financial Support Scheme (CIFS) covered through September 2010 all retail and corporate deposits (excluding retail deposits covered by the deposit insurance scheme), interbank deposits, covered bonds, senior unsecured debt, and certain subordinated debts. The guarantee was equivalent to 240 percent of GDP.

7

See the exchange of letters between the ECB President and the Irish authorities at http://www.ecb.europa.eu/press/html/irish-letters.en.html

8

See The Chairman’s Summing Up: Ireland—Request for an Extended Fund Arrangement (BUFF/10/174, 10/16/10); and Ireland—Request for an Extended Arrangement (EBS/10/223, 12/4/10).

9

See Ireland: Request for an Extended Arrangement—Staff Report; Staff Supplement; Staff Statement; and Press Release on the Executive Board Discussion (IMF Country Report No. 10/366, 12/4/10).

10

See Ireland—Request for an Extended Arrangement—Staff Report; Staff Supplement; Staff Statement; and Press Release on the Executive Board Discussion (IMF Country Report No. 10/366, 12/4/10), pp. 5, 29-30 and Box 1; and The Chairman’s Summing Up: Ireland—Request for an Extended Fund Arrangement (BUFF/10/174, 10/16/10).

11

See Ireland—Request for an Extended Arrangement (EBS/10/223, 12/4/10); and Ireland—Assessment of the Risks to the Fund and the Fund’s Liquidity Position (EBS/10/223, Supplement 2, Rev. 1, 12/7/10).

12

A brief reference to euro area issues was included in EBS/10/223, Supplement 2, op. cit.

13

It was agreed to defer the required completion of capital injections until after the General Election, to let the new government decide on the recapitalization. The recapitalization of the banks took place in July 2011, following the PCAR.

14

These reforms were subject to detailed structural conditionality under the EC Memorandum of Understanding. A summary was included in the Memorandum of Economic and Financial Policies with the Fund.

15

See “Ireland—Tenth Review Under the Extended Arrangement” IMF Country Report No. 13/163 (6/19/13), Box 1.

16

The patent cliff had, however, little effect on domestic resident income and average GNP growth was higher-than-programmed (1.5 percent per annum versus 0.2 percent projected at the time of the program request).

17

Even when adjusted for estimates of the output gap, Ireland’s current account position strengthened, also in comparison with some other countries in the euro area periphery; see J.S. Kang and J.C. Shambaugh, 2014, Progress Towards External Adjustments in the Euro Area Periphery and the Baltics, IMF Working Paper (WP/14/131).

18

See also Euro Area Policies —Selected Issues chapter on deleveraging, (SM/13/206, 7/12/13).

19

The PCAR covered AIB, BOI and PTSB, labeled “intervened banks” in the rest of the paper, as well as EBS Building Society, which was merged with AIB in 2011.

20

Under European Union State Aid rules, banks were also required to undergo a forced restructuring and submit a restructuring plan within six months of receiving government support. While BoI’s and AIB’s plans were approved, approval of PTSB’s restructuring plan is still pending.

21

The 2011 PCAR loan losses and PLAR deleveraging plans were initially prepared with the assumption that a second NAMA was going to take place for larger land and development loans of less than €20 million. However, the authorities ultimately decided to have banks retain these assets. See Box 1 in PCAR 2011 Review: Analysis of PCAR banks up to end June 2012 Compared to PCAR 2011 at www.centralbank.ie.

22

The PCAR resulted in capital requirements of €24 billion, including a contingent equity buffer of €3 billion.

23

This number does not include the transfer of banking assets to NAMA, as these took place before the program started. Including such transfers would show a contraction in assets by more than one third.

24

The CBI indicated to banks that, effective January 1, 2014, if it found insufficient progress on meeting restructuring targets of NPLs on a sustainable basis, it would use its regulatory powers to impose Pillar II requirements, including holding additional capital or apply specific provisioning requirements.

25

Some €0.5 billion of additional fiscal measures were taken in 2012 to meet the deficit target, but were largely unwound by somewhat lower measures in 2014.

26

The CBI set up a separate loan workout resolution framework for SMEs. SMEs also had recourse to corporate insolvency proceedings where the legal framework was generally considered more effective.

27

For example, the insolvency framework in Norway (introduced in 1993 following a banking crisis and housing market bust) allows for both a voluntary and a compulsory court administered debt settlement where in the latter adjustments are made to reduce the loan-to-value ratio to a maximum of 110 percent (for further details, see Box 3). Denmark, Finland, and Sweden have broadly similar systems in place. For a broader discussion, see Y. Liu and C. Rosenberg, Dealing with Private Debt Distress in the Wake of the European Financial Crisis—A Review of the Economics and Legal Toolbox (2013), IMF Working Paper 13/44.

28

PTSB is weighed down by a large portfolio of low yielding tracker mortgages (some two-thirds of its total assets in 2013), which have remained on the bank’s balance sheet. Notwithstanding considerable improvements in its management, the government’s 2012 restructuring plan acknowledged that PTSB’s asset-liability structure offered little prospects for profitability over the medium term under reasonable funding cost assumptions.

29

A key objective of resolving a banking crisis is to create viable financial institutions. Recapitalization alone is not a sufficient condition to secure financial viability of a bank over the medium term. The bank restructuring plan should also address the underlying problems in the bank’s business practices that led to the weaknesses and its recapitalization needs; see IMF, 2002, Building Strong Banks through Surveillance and Resolution; and BCBS, 2002, Supervisory Guidance on Dealing with Weak Banks. Viability is also required under the EC’s Guidelines on Restructuring Aid to Banks, which requires that the government submits a restructuring plan that demonstrates how the bank will restore long-term viability without state aid as soon as possible.

30

It is notable, however, that the Irish banking sector is contestable for foreign institutions and, indeed, some foreign presence remains in the retail market and other foreign banks operated in Ireland in the past.

31

See Ireland—Twelfth Review Under the Extended Arrangement and Proposal for Post-Program Monitoring (EBS/13/151, 12/03/13).

32

Subordinated debt was not fully written down, as some cash payouts were provided to subordinated debt holders.

33

See, FSB, 2011, Key Attributes of Effective Resolution Regimes for Financial Institutions.

34

See Ireland—Twelfth Review under the Extended Arrangement and Proposal for Post-Program Monitoring (IMF Country Report No. 13/366), Annex I, December 2013.