Greece: Ex-Post Evaluation of Exceptional Access Under the 2012 Extended Arrangement

In accordance with Fund policies, this report conducts an ex-post evaluation of a four-year exceptional access extended arrangement under the Extended Fund Facility (EFF) with Greece approved in March 2012. The Fund committed €28 billion under the extended arrangement (SDR 23.8 billion or 2,159 percent of Greece's quota at the time), following the cancellation of the 2010-12 Stand-By Arrangement (SBA). The program was supported by Greece's EU partners, who committed €144.7 billion. Significant private sector debt relief (€106 billion) was completed at the outset of the program and large official debt relief was provided as well. The Fund disbursed SDR 10.2 billion. Only five out of 16 program reviews were completed as the program went off track finally in mid-2014. The arrangement was cancelled in January 2016.

Abstract

In accordance with Fund policies, this report conducts an ex-post evaluation of a four-year exceptional access extended arrangement under the Extended Fund Facility (EFF) with Greece approved in March 2012. The Fund committed €28 billion under the extended arrangement (SDR 23.8 billion or 2,159 percent of Greece's quota at the time), following the cancellation of the 2010-12 Stand-By Arrangement (SBA). The program was supported by Greece's EU partners, who committed €144.7 billion. Significant private sector debt relief (€106 billion) was completed at the outset of the program and large official debt relief was provided as well. The Fund disbursed SDR 10.2 billion. Only five out of 16 program reviews were completed as the program went off track finally in mid-2014. The arrangement was cancelled in January 2016.

Introduction

1. This report conducts an Ex-Post Evaluation of Greece’s 2012-16 extended arrangement under the Extended Fund Facility (EFF). The Fund committed €28 billion under the EFF (SDR 23.8 billion or 2,159 percent of Greece’s quota at the time), following the cancellation of the 2010–12 Stand-By Arrangement (SBA).1 Similarly to the SBA, the Fund closely collaborated with its Troika partners—the European Commission (EC) and the European Central Bank (ECB). The Fund disbursed SDR 10.2 billion with five out of 16 reviews completed before the arrangement was cancelled in January 2016. The EFF was an exceptional access arrangement, which requires an ex-post evaluation to (i) review performance against original program objectives; (ii) discuss whether program design was appropriate to address Greece’s challenges; and (iii) assess whether program modalities were consistent with Fund policies.2

2. In the wake of the euro adoption, Greece enjoyed rapid but unsustainable income gains. The availability of easy financing following the euro adoption in 2001 allowed the public sector to borrow extensively, on average 8 percent of GDP per year in 2002–09 with public debt reaching 127 percent of GDP by 2009. In addition, starting in the mid-2000s, the private sector began to rapidly accumulate external debt, which rose to about 175 percent of GDP by 2009. A massive fiscal impulse (28 percent of GDP in 2002–09) and private borrowing generated strong growth, resulting in a rapid but unsustainable nominal income convergence with higher income countries in the EU and significant real effective exchange rate appreciation (Figure 1).3 The strong economic growth, however, masked the reality of low investment, poor use of labor and capital, a large and inefficient public sector, rising vulnerabilities in the banking system, a cumbersome judicial and legal system, and widespread informality (McKinsey & Company 2012).

Figure 1.

Greece: Imbalances and Convergence

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: European Commision; Haver Analytics; national statistics office; and IMF staff calculations.

Pre-program: SBA 2010–12

3. The SBA with access of €30 billion (SDR 26.4 billion)4 was approved in May 2010 and cancelled in March 2012. The European partners provided strong financial support for the program by committing €80 billion.5 The program was successful in achieving strong fiscal consolidation and implementing labor market reforms. However, the recession was deeper than expected, financial sector fragilities increased, and debt sustainability was not achieved.

4. The accumulated imbalances began to be corrected by way of a deep recession during 2010–11. As a result of the sudden stop in private capital inflows (Figure 2), collapsing confidence, a sizable fiscal consolidation, and credit contraction, real output declined by almost 18 percent by 2011 from its pre-crisis peak in 2007, and the unemployment rate approached 18 percent.

Figure 2.
Figure 2.

Greece: Gross Capital Flows

(Billions of USD)

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Source: Financial Flows Analytics database.

5. Greece undertook a significant fiscal adjustment under the SBA during 2010–11. The cumulative improvement in the primary fiscal balance during this period was about 7 percentage points of GDP, but the primary fiscal balance remained in deficit (2.4 percent of GDP in 2011),6 the overall fiscal deficit stood at more than 10 percent of GDP in 2011, and debt ended on an unsustainable path. The adjustment was mainly revenue-based on the back of various measures, including VAT rate hikes, a new property tax, and income tax base-broadening measures.7 Investment, other discretionary expenditures, and to a lesser extent wages, bore the burden of adjustment on the expenditure side. Owing to falling GDP, relatively modest decreases in primary expenditure-to-GDP ratios masked substantial nominal expenditure cuts (Figure 3).

Figure 3.
Figure 3.

Greece: Fiscal Outcomes under the SBA

(Percent of GDP)

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Source: IMF staff reports.

6. Debt restructuring became inevitable. The possibility of debt haircuts loomed larger after the Deauville Summit of October 2010 where it was envisaged that future crisis resolution would require an “adequate participation of private creditors.”8 The plan for a private sector involvement (PSI) was first announced in July 2011, and the completion of the debt exchange, involving a haircut on private sector creditors of 53 percent, would become a prior action for the March 2012 EFF request (Box 2). Limited debt relief from official creditors (Official Sector Involvement—OSI) was provided as well in the run-up to EFF approval (Box 2).

7. Greece continued to face large external imbalances and low competitiveness during the SBA period (Figures 4 and 5). In 2011, the current account deficit was close to 10 percent of GDP. While unit labor costs declined by 9 percent between March 2010 and March 2012, consumer prices did not follow—mostly owing to significant structural rigidities and VAT increases. Structural factors and non-price competitiveness were constraining export performance further. Greek exports are concentrated in low- and medium-technology sectors in relatively small enterprises, making it difficult to scale them up in response to potentially beneficial changes in relative prices (Athanasoglou et al. 2010). The overall business environment was not conducive to entrepreneurship: Greece ranked 100th in the World Bank’s 2012 Doing Business Indicator, and 90th in the World Economic Forum’s Global Competitiveness Index. Also, the onset of a recession in the euro area in 2011, where most of Greece’s trading partners are located, further weighed on prospects for an export-led recovery. As a result of the above-mentioned factors, Greece’s export market share steadily declined from 2008 onward (OECD 2012).

Figure 4.
Figure 4.

Greece: Competitiveness and Export Performance

(Index, 2007Q1=100)

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Source: OECD.1/ Growth of exports relative to the growth of the country’s export market. See Le Fouler et al. (2001).
Figure 5.

Greece: Exports: Composition and Destinations, 2011

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Source: The Atlas of Economic Complexity.

8. Banks came under severe solvency and liquidity pressures due to the PSI and Grexit fears. The proposed PSI exposed the banks to potential losses on their holdings of government debt, which stood at more than twice the banks’ Tier 1 capital. At the same time, currency redenomination fears led to deposit and funding outflows. As a result, in the course of 2011, customer deposits fell by 17 percent, foreign bank funding contracted by one-third, the bank equity index collapsed, and the CDS spreads on subordinate bank debt were signaling an imminent default.

9. Vulnerabilities related to private sector indebtedness continued to grow (Figure 6). Rapid credit growth and large increases in property prices pre-2008 contributed to a buildup of the systemic risk. Several years of economic contraction weakened the banking system’s asset quality, with the ratio of non-performing loans (NPLs) doubling to 20 percent between 2010 and 2011. The Greek authorities commenced a foreclosure moratorium and debtor-friendly reform of household insolvency rules (the so-called “Katseli Law”) in 2010, which contributed to undermining payment culture of borrowers.

Figure 6.

Greece: Selected Financial Sector Indicators

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: BIS; IMF World Economic Outlook; and SNL.1/ NPLs include impaired restructured loans.

10. Ownership of the program supported by the SBA turned out to be limited. Under the SBA, negotiated and implemented by the center-left Pasok government, Greece achieved significant fiscal adjustment. But structural reforms started to face significant opposition from vested interests, and popular discontent was rising, with no strong pro-reform sentiment in the society at large. In October 2011, Prime Minister (PM) Papandreou (of the center-left Pasok) announced a referendum on the rescue package negotiated with the Troika, only to withdraw his referendum plan a month later.

11. Vulnerable countries in the euro area experienced significant market pressures in late 2011–early 2012 (Figure 7). While joint EC-ECB-IMF programs were in place in Ireland and Portugal, the yields on their government securities were close to record highs.

Figure 7.
Figure 7.

Ten-year Sovereign Bond Yields

(Percent)

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Source: Bloomberg.

Policy Trade-Offs and Program Objectives

A. Key Policy Trade-Offs

12. Because the discussions on the EFF unfolded at the time of exceptional fragility in Greece and in the euro area, the Greek authorities and their European partners were facing several acute policy trade-offs:

  1. How to strike a balance between adjustment and financing. Even with the unprecedented PSI, a further significant fiscal adjustment was unavoidable because of the large scale of imbalances and limited availability of official sector debt relief. The adjustment had to come, however, in the wake of the sharp output decline and rising social pressures. Additional financing could dull the pain of adjustment, but at a cost of raising the high stock of public debt (post-PSI) further. Moreover, there were political constraints on the amount of financing that Greece’s EU partners could provide, in part reflecting concerns that these countries were being asked to finance pensions and other expenditures in Greece that were viewed as excessive relative to other member states. As a result, a lengthy and difficult process was required for EU member states to reach agreement on the financing envelope. In addition, Greece’s European partners insisted that Greece adhere to the euro area’s common economic framework under the EU Stability and Growth Pact as soon as possible.

  2. How to achieve external adjustment and improve competitiveness. Competitiveness and external balances could be improved either by an outright currency devaluation or by an internal devaluation. The former would be quick but highly disruptive, requiring an exit from the euro area, a likely default on public and private sector debt, and a period of legal uncertainty because the euro exit process had not been codified. The latter would take time and require comprehensive and politically difficult structural reforms.

  3. How to reach an acceptable solution in time to avoid spillovers. While temporary firewalls were already in place in the euro area, they were by no means perceived as adequate in particular for larger countries. A disorderly exit and/or default could have easily destabilized vulnerable countries and, in the extreme case spread to the core of the euro area. Therefore, timely agreement on a program was essential.

13. The leadership of the main political parties in Greece was coalescing behind commitment to adjustment and reform in the wake of EFF discussions. Following the G-20 summit in Cannes (November 2011), Grexit became a real possibility, as further EU support to Greece was clearly made conditional on progress with reforms. This catalyzed the formation of a technocratic coalition government, including Pasok (center-left party) and New Democracy (center-right party), in November 2011, which appeared to have cemented the emerging political consensus for reforms in support of Greece’s choice to remain in the euro area. The newfound domestic consensus raised hopes that the implementation challenges experienced under the SBA could be overcome.

B. Program Objectives

14. The EFF’s broad objectives were to restore competitiveness and growth, fiscal sustainability, and financial stability (IMF 2012). With the encouragement and endorsement of their EU partners, the Greek authorities committed to rapid fiscal and external adjustment to be accompanied by politically difficult structural reforms in support of their strong preference to remain in the euro area.9 “The authorities argued that prolonging the adjustment path beyond 2014 would pose risks to credibility and, given resistance from their European partners, worried that this would be seen as a lack of commitment to Stability and Growth Pact targets” (IMF 2012). Political assurances of policy commitment were secured from the main political parties. At the same time, staff argued that “demand effects from the implementation of structural reforms, as well as weaker economic prospects in Europe, called for a longer adjustment period (thus also allowing a more accommodative fiscal policy in the near term)” (IMF 2012). Nevertheless, staff felt that the Greek authorities should be given a chance to implement their preferred policy mix. Staff candidly explained the political and economic risks to the program: ”… the new program is a bold step in the right direction but given the challenges and the track record is subject to exceptional risks” (IMF 2012).

15. Competitiveness would be restored by a relatively large and rapid internal devaluation, supported by an ambitious set of labor market reforms (to bring wages in line with productivity and reduce unit labor costs (ULC) by 15 percent) and product and service market reforms (to translate wage competitiveness into price competitiveness). Privatization was expected to present new investment opportunities and productivity gains, thereby supporting employment and growth. Exports were projected to increase by 17 percent and imports to decline by 3 percent from 2011 to 2015. As a result, the current account deficit was projected to decline to about 3 percent of GDP by 2015, from about 10 percent in 2011.

16. Restoration of confidence and a relatively quick payoff from structural reforms were assumed to underpin economic recovery. The economy was projected to contract by almost 5 percent in 2012, in part reflecting a fiscal drag and falling wages. Real GDP growth was projected to be zero in 2013 and to increase to 2.5–3.0 percent over the medium term, as exports and private investment were expected to recover significantly on account of improved investor sentiment and benefits of reforms. The growth rate was projected to eventually converge with the long-run potential growth of 1-1½ percent. Unemployment was projected to peak at 19 percent in 2013.

17. To reduce the stock of public debt to 120 percent of GDP by 2020 (Figure 8), the EFF envisaged an ambitious fiscal adjustment supported by a large privatization program. To meet that objective under the agreed-upon financing envelope and debt relief, the targeted fiscal adjustment had to be large and fast-paced, with a seven-percentage-point-of-GDP improvement in the primary fiscal balance—comparable to what had already been attained during 2010–11—to be achieved in the course of three years (Table 1).10 The adjustment would largely rely on primary expenditure cuts (almost 8 percentage points of GDP), primarily in social benefits, including pensions, and wages, although most of the expenditure cuts in 2013 and beyond were expected to come from hitherto unidentified measures. Revenue measures were to focus on equity, efficiency, and enforcement considerations.11 Privatization proceeds from real estate, financial, and other assets were projected at €46 billion during 2012–20 (unchanged from the last review under the SBA) and were expected to reduce public debt by 20 percent of GDP by 2020. A comprehensive fiscal structural reform agenda was agreed in support of the targeted fiscal adjustment (Table 2).

Figure 8.
Figure 8.

Greece: Debt Evolution

(Percent of GDP)

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Source: IMF staff reports.
Table 1.

Greece: Program Fiscal Objectives under the EFF (2011–15)1/

article image
Sources: IMF staff reports and estimates.

At the time of the EFF request in 2012.

Unidentified measures were assumed to be on the expenditure side at the time of the EFF request.

Table 2.

Greece: Focus of Fiscal Structural Reforms under the EFF

article image

18. With bank capital wiped out by the PSI operation, and the recession weighing on asset quality, recapitalization and resolution became key. Of the €50 billion earmarked for bank recapitalization costs, one half was allocated to offset the expected PSI-related losses, and the other half—equivalent to 5 percent of banking system assets—was made available to resolve problem banks and deal with the existing and future credit losses. The program envisaged improving supervision and regulation, as well as establishing a stronger governance framework for banks recapitalized with public funds. As the program progressed, strengthening private debt restructuring frameworks and banks’ NPL management capacity were identified as higher priorities.

C. EU Support

19. European partners committed large financial support in the context of their program with Greece. The European Financial Stability Fund (EFSF) program’s commitment of €144.7 billion covered the period of March 2012–June 2015.12 The EFSF program contained structural conditionality and significant capacity building components. The discussions on the EFF and the EFSF programs were closely coordinated in the context of the Troika arrangements. The ECB showed willingness to provide exceptional liquidity support to Greek banks. Limited debt relief in the context of the OSI was also provided at the beginning of the program (Box 2).

Program Outcomes

A. Timeline and Political Context

20. Contrary to the assumption of broad political support for the program and the related positive confidence effects, political turbulence prevailed during the program period. The technocratic administration of PM Papademos successfully negotiated the EFF and debt relief, only to be removed from office shortly thereafter. What followed was a period of social pressures and political crises (Table 3). Greece had six Prime Ministers and nine Finance Ministers during the EFF, with some serving as little as a few weeks. The continued deep recession put a severe strain on the social fabric of the country (Figures 9 and 10). Frequent episodes of political instability and the failure of the Greek political system to deliver a broad-based pro-reform coalition, which became increasingly evident, created exceptionally difficult conditions for program implementation.

Table 3.

Greece EFF: Chronology of Key Events

article image
Figure 9.
Figure 9.

Greece: Real GDP Growth and Unemployment Rate

(Year-on-year percentage change; quarterly data)

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: Hellenic Statistical Authority; and Haver Analytics.
Figure 10.
Figure 10.

Selected Social Indicators

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: Eurostat; national statistics offices; and Haver Analytics.

21. Program implementation was strongly correlated with political developments. During a tumultuous 2012 with two parliamentary elections and intense Grexit fears, program implementation was poor. As a result, the assumed confidence effects did not take hold in 2012. There was a period of relative stability during early 2013–early 2014 with stronger program implementation reflected in over-performance on fiscal targets. By mid-2014, green shoots of recovery were clearly perceptible, and the Greek government was able to return to debt markets. However, after a deeper-than-expected 5-year recession and a 30-percent fall in real disposable incomes, the rise of popular discontent was unstoppable. As a result, the parties opposing the agreed adjustment program gained ground throughout this period, culminating in the collapse of support for the traditional political parties and the Syriza government assuming power in early 2015. In the first half of 2015, both public and official opposition to adjustment and reforms ran strong, leading to an overwhelming rejection of a proposed EU-supported program by the Greek voters in the July 2015 referendum. The resulting interruption of the financial support to the sovereign and of the liquidity support to Greek banks against the background of renewed Grexit fears necessitated a temporary closure of banks and the imposition of deposit withdrawal limits and capital controls. In the midst of the crisis, Greece became the first advanced country to accumulate overdue financial obligations to the Fund between June 30 and July 20, 2015.

22. The initial phasing of the program had to be revamped significantly. The initial program envisaged 16 quarterly reviews. However, only five reviews (of which two combined reviews) were completed with significant delays during March 2012–June 2014, as significant implementation and macro risks materialized. After June 2014, the program went irretrievably off track. The staff continued discussions with the authorities and provided some TA from June 2014 till the cancellation of the program in January 2016. In July 2015, staff updated and published a DSA that concluded that public debt was unsustainable. A blog post by the IMF Chief Economist at the time, Olivier Blanchard, explained the way forward for the Fund: “The role of the Fund in this context is not to recommend a particular decision, but to indicate the tradeoff between less fiscal adjustment and fewer structural reforms on the one hand, and the need for more financing and debt relief on the other” (Blanchard 2015).

23. Economic outcomes for part of 2014 and the entirety of 2015 cannot be fully attributed to program design because the program was off track during this period. Some program measures that were implemented during 2012–14 continued to impact the economy afterwards, but for the most part, the outcomes of the last 18 months of the program were attributable to the inconsistent policies pursued by the Greek authorities outside the program context, including backtracking on a number of important policies adopted under the EFF. The discussion below regarding outcomes during the program period is subject to this important caveat.

B. Real and Nominal Growth

24. Growth fell short of original program expectations and unemployment was well above program projections during the entire period under review. The recession was much deeper than originally expected during 2012–13, and the nascent recovery of 2014 stalled with the escalation of political tensions. Real growth turned negative again in 2015. Deflators were below projections except in 2012. As a result of lower growth and lower deflators, nominal GDP in 2013 and 2015 was about 12 and 20 percent below the original forecasts, respectively. Unemployment was much higher than projected under the program. The non-seasonally adjusted rate had peaked at about 29 percent in 2013 and then gradually declined to about 25 percent in 2015.

25. Multiple factors explain growth underperformance. The risks of political turmoil and of incomplete and inconsistent program implementation materialized, undermining investor confidence and growth. Furthermore, repeated downgrades to the WEO outlook for the euro area were an important factor. However, the initial macroeconomic assumptions of the program may have been too optimistic. The short- and medium-term growth payoffs of reforms were in a highly optimistic range,13 fiscal multipliers may have been underestimated,14 and the possible negative feedback loops stemming from the rapidly rising insolvency problems in the private sector15 and the persistence of Grexit fears16 were not fully factored in. In addition, the assumed strong positive impact of improved confidence on investment growth (largely through retained earnings of exporters) appears too optimistic in light of the challenges facing Greek exporters (para. 30–31) and the financial system (para. 37–39), as well as the large debt overhang in the public and private sectors.17 Also, political factors may have not been fully exogenous, as the protracted recession and the rapidly falling living standards may have undermined both political support for the program measures and investor confidence.

26. The program had to be substantially re-designed during the combined first/second reviews in early 2013 (Figure 11). This reflected expectations of slower reform progress, substantially weaker investment, and lower payoffs from reforms. Also, fiscal multipliers were raised. These factors led to a much more pessimistic growth forecast. Furthermore, nominal GDP was adjusted down by 3 percent on account of data revisions. In addition to the real growth downgrade, deflators had to be adjusted downward as well. Consequently, by 2015 projected nominal GDP was 12 percent lower than envisaged in the program request. With much lower projected nominal GDP, the initial fiscal objectives became less achievable, warranting a reduction in the primary balance targets, additional fiscal measures, and more debt relief in the context of the OSI. Moreover, the revised program framework accounted for a stronger-than-expected external adjustment, and the current account was projected to be in balance by 2015.

Figure 11.
Figure 11.

Greece: Macroeconomic Projections and Outcomes

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: IMF Staff Reports; Haver Analytics; and IMF staff calculations.

C. Fiscal Adjustment

27. During 2012–15, fiscal adjustment was significant (3¼ percent of GDP) but fell short of the program objective (7 percent of GDP). The primary fiscal balance improved from a deficit of 3 percent of GDP in 2011 to a surplus of ½ percent of GDP in 2013, exceeding the revised program target. Once the program went off track, the primary surplus fell to ¼ percent of GDP by 2015, below the initial program target of 4.5 percent of GDP and the revised program target of 3.5 percent of GDP. Privatization proceeds were €1.9 billion during the program period, representing about 10 percent of the initial program target.

28. In terms of the composition of fiscal adjustment, the objectives set out in the EFF request were not met (Figures 12a for the period 2012–15 and 12b for the period 2012–14). Staff argued for more equitable taxation (including better enforcement and a broadening of the tax base), improvements in the social safety net, and more sustainable wage and pensions expenditures. In contrast, the actual adjustment on the revenue side (3 ¾ percent of GDP) during 2011–15 largely focused on mostly ad hoc measures, including regressive and distortionary taxation on small bases as revenue administration remained fundamentally weak. Primary expenditure increased slightly (½ percent of GDP) with pensions expenses rising significantly relative to GDP (para. 29).18 Regarding the social safety net, staff argued for a targeted guaranteed minimum income program, a pilot for which was rolled out only in early 2015. Despite significant misgivings of the staff, this sub-optimal adjustment mix was accommodated in program reviews.

Figure 12a.
Figure 12a.

Greece: Fiscal Objectives and Outcomes, 2011–15 1/ 2/

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: IMF staff reports and calculations.1/ Request: the difference between the 2015 outcome projected at the time of the EFF request and the 2011 outturn estimated at the time of the EFF request. Actual: the difference between the 2015 and 2011 actual outturns estimated in 2016.2/ Actual data are based on ESA2010, and in program definition; whereas EFF request data are based on ESA1995. Request data includes unidentified measures assumed at the time of the EFF request.
Figure 12b.
Figure 12b.

Greece: Fiscal Objectives and Outcomes, 2011–14 1/ 2/

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: IMF staff reports and calculations.1/ Request: the difference between the 2014 outcome projected at the time of the EFF request and the 2011 outturn estimated at the time of the EFF request. Actual: the difference between the 2014 and 2011 actual outturns estimated in 2016.2/ Actual data are based on ESA2010, and in program definition; whereas EFF request data are based on ESA1995. Request data includes unidentified measures assumed at the time of the EFF request.

29. Progress on fiscal structural reforms was limited:

  • In PFM, a measure of progress was achieved, such as better fiscal reporting, a clear assignment of financial management responsibilities and more effective processes in the ministry of finance, and improved cash management operations, including the establishment of a treasury single account for the central administration. Progress, however, was lacking in other areas, including staffing, modernizing payment processes, and halting accumulation of spending arrears (Box 3).

  • In revenue administration, although some compliance initiatives have been successfully implemented, fundamental weaknesses in core operations remain, and as a consequence overall progress on enforcement and compliance was limited, in part reflecting still-insufficient autonomy of the General Secretariat for Public Revenues (GSPR) due to political interference (Box 4).

  • As part of the pension reform, immediate overruns were not contained despite cuts in pension benefits and a reduction of the number of payments per year from 2012. The long-term sustainability was not addressed, either, with the deficit of the pension system amounting to 11 percent of GDP at end-2015, the highest level in the EU (Box 5).

  • In tax policy, the new property tax ENFIA was introduced but reforms on property valuation stalled. Backtracking occurred with the implementation of the 2013 income tax code (which streamlined tax rates and tax expenditures), as some policies introduced originally were reversed and a generous tax credit was introduced (Box 6).

D. External Adjustment

30. The external current account adjusted rapidly. Despite the weaker fiscal adjustment, the external current account balance in 2015 turned positive, exceeding the original program objective. Most of the adjustment was due to falling imports, which declined by nearly one fifth between 2011 and 2015, while exports remained broadly stable. The ULC-based REER declined by about 25 percent, while the CPI-based REER declined by about 10 percent between 2011 and 2015 (Figure 13). This difference is in part explained by persistent rigidities in product and service markets. Even though it is uncertain whether the planned product and service market reforms (if fully implemented) could have reduced these rigidities with a positive impact on competitiveness during the program period, insufficient implementation progress on these reforms (para. 33–34) negatively affected medium- and long-term prospects for improving competitiveness.

Figure 13.
Figure 13.

Greece: REER and Export Performance

(Index, 2010=100)

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: IMF INS database; and OECD.1/ Growth of exports relative to the growth of the country’s export market.

31. Exports underperformed, despite the significant adjustment in labor costs (Figure 14). By 2015, exports did not exceed their 2008 peak, and Greece’s market share did not improve during the program period.19 Delays and inconsistent implementation of program reforms and weaker-than-envisaged activity in the euro area contributed to the weak export performance.20 Also, a relatively low share of easily scalable tradable outputs, severe liquidity constraints (which were not foreseen at the program inception) and lack of investment impeded the re-allocation of resources in support of export growth.

Figure 14.

Greece: Export Performance and Competitiveness

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: IMF staff reports; World Bank; and World Economic Forum.1/ Actual numbers are in BPM6.

E. Structural reforms

32. Initially, progress was achieved on a number of structural reforms. At the program’s outset, collective bargaining was made less rigid, the setting of the minimum wage reformed, and—with a view to liberalizing the regulated professions and product markets—the relevant legislation was screened and amended. In the course of the program, the authorities undertook steps to strengthen the framework for privatization, reform barriers to competition, create a more business-friendly environment, and make administration of justice quicker and less costly.

33. However, reform momentum weakened, as further program implementation faced strong opposition from vested interests. Even though labor market reforms bore fruit early on—with a noticeable decline in unit labor costs—significant restrictions regarding collective dismissals, industrial actions, and the setting of minimum wages remained in place. Moreover, the authorities began reversing these reforms (by restoring temporarily the pre-program collective bargaining framework and then re-instating some of the aspects of the program framework, among other steps) even before the EFF was cancelled. Product market reforms—including opening up the regulated professions—were noticeably less successful, with the legislation passed but not meaningfully implemented. By the time of the 5th review, some results were achieved in reducing case backlog in administrative courts, but implementation stalled after that.

34. With only limited improvements, Greece failed to catch up with other euro area countries. There have been improvements in a few areas (for instance, the ease of starting a business and trading), but in others—such as registering property, access to credit, and enforcing contracts—the distance to best practice has increased (World Bank 2016). The country still scores low in key areas of contract enforcement, insolvency resolution, and access to credit, and its ranking in the Global Competitiveness Index has not improved significantly (Schwab 2016). Moreover, Greece stands out among other euro area program countries for the poor quality of its institutions prior to the crisis and a further deterioration in the governance indicators during the program (Alcidi et. al. 2016).

F. Debt Sustainability

35. PSI and OSI were exceptionally large by international comparisons,21 but they achieved a relatively modest immediate decline in the stock of public debt. Public debt declined to €305 billion (160 percent of GDP) at end-2012, from €356 billion (172 percent of GDP) at end-2011. This is explained by a number of factors. First, Greece’s EU partners excluded from the debt restructuring the Greek bond holdings by EU institutions (i.e., the ECB, national central banks, and the EIB), which amounted to more than 16 percent of total public debt. Second, domestic banks represented around 40 percent of the bond holders (or 24 percent of total public debt), and the resulting PSI-related losses needed to be debt financed. Third, the terms of the bond exchange had to be attractive (e.g., acceptable haircut and near-cash sweetener) to ensure wide participation, as Greece’s EU partners saw significant merit in a voluntary exchange to avoid contagion. Fourth, there were constraints on the OSI related to existing official EU loans (15 percent of total public debt). In light of the large amount of official financing already given to Greece, there was strong political resistance in some EU countries on outright reductions of official debt (para. 12). Fifth, the delay of debt restructuring reduced private bond holdings subject to restructuring, as amortization payments to the private sector amounted to about €50 billion during 2010–early 2012. Sixth, nominal GDP declined by almost 8 percent in 2012.

36. Despite the PSI and OSI, and some progress in fiscal adjustment, debt was assessed as unsustainable in June and July 2015 (Figure 15). When the program was on track, at the time of all completed reviews, debt was assessed as sustainable but not with high probability. This assessment was predicated on the assumptions that Greece could maintain fiscal primary surpluses in a 3.5–4.5 percent of GDP range in the medium and long terms and that Greece’s European partners would follow through with their commitment to further support Greece if needed as long as the program was implemented. After the program had gone off track, the published DSAs (IMF 2015a and IMF 2015b) were adjusted based on important lessons learnt from EFF implementation. The macro framework was overhauled, with lower growth, primary fiscal surpluses, and privatization proceeds22 over the medium and long terms. Attaining a debt level of 120 percent of GDP by 2020 was deemed no longer possible (Figure 15). In light of the highly concessional nature of Greece’s debt, the 2015 DSAs appropriately suggested reframing the sustainability assessment around the gross financing needs (GFN) metric rather than a specific debt level. The GFN benchmark to be used would be 15-20 percent of GDP, consistent with the Fund’s framework for debt sustainability analysis for market-access countries (MAC DSA). To attain this new goalpost, yet more debt relief from the European partners would be needed.

Figure 15.
Figure 15.

Greece: Gross Debt and Financing Needs 1/

(Percent of GDP)

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Source: IMF’s June 2015 Debt Sustainability Analysis.1/ Projections without concessional financing.

G. Financial sector

37. Despite progress in banking system consolidation, the financial system remained vulnerable, and banks were poorly positioned to support economic expansion by end-2015 (Figures 16 and 17). Between 2012 and 2015, banks had some success in cutting costs and increasing efficiency, with the number of bank branches reduced by a third and the number of staff by nearly a quarter, which is in the top quintile of declines seen in the European Union during that period. Three rounds of recapitalization, totaling some €68 billion (€85 billion including deferred tax credits) brought the Tier 1 capital adequacy ratio to about 15 percent by end-2015. However, Greek banks still suffered from fragile balance sheets (an NPL ratio of 45 percent) and extraordinary dependence on central bank funding at end-2015. The bank-sovereign nexus continued to be a problem due to the banks’ exposure to the Greek government via holdings of deferred tax credits.23

Figure 16.
Figure 16.

Greece: Tier 1 Capital Ratio 1/

(Percent)

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Source: IMF staff calculations.1/ Public sector funds for recapitalization were commited in 2012 but disbursed over 2012 and 2013.
Figure 17.

Greece: Selected Banking System Indicators

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: Bank of Greece; Haver Analytics; and IMF staff calculations.1/ Includes €7.2 billion of resolution costs for non-core banks in 2012; €9.2 billion in 2013.2/ Liability management exercise: net assets created through buyback of debt and hybrid capital instruments below par.3/ Data for 2013 show increase in deferred tax assets eligible towards Core Tier 1 equity due to Bank of Greece’s removal of the 20 percent prudential filter; deferred tax credits introduced beginning in 2014.

38. Credit to the private sector has continuously contracted since 2011, contributing to the poor growth performance (Figure 18). The contraction of credit, still evident in 2015, was particularly relevant for SMEs, which play a large role in the Greek economy. Credit demand was weak during the entire duration of the program, with the temporary exception of 2014, in line with the economic recovery. On the supply side, banks almost continuously tightened their lending standards (with the exception of 2014), making access to credit more difficult and therefore amplifying the economic contraction. Chances for a creditless recovery in Greece are uncertain because its economy is dominated by credit-dependent SMEs.

Figure 18.
Figure 18.

Greece: Credit Developments

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: ECB (Bank Lending Survey); IMF staff reports and calculations.1/ Positive values indicate tightening lending standards and higher credit demand.

39. With the benefit of hindsight, the program could have done more to address several potential vulnerabilities in a timely manner. Bank capital needs assessment exercises could have used more conservative estimates for the scale and severity of credit losses (Figure 19), reducing the need for subsequent re-capitalizations.24 Furthermore, while certain measures on private debt restructuring and NPLs management were introduced early in the program, a comprehensive strategy to tackle NPLs and insolvency frameworks was only adopted relatively late (during the 5th Review) and largely in reaction to poorly-designed government initiatives,25 when rising household and corporate bankruptcies made insolvency reform even more politically difficult. Finally, while conditionality on governance of the banks and of the state-owned recapitalization vehicle (the Hellenic Financial Stability Fund, or HFSF) was nominally met after the PSI, close links between the senior leaders of the banks, political parties, and large corporations were not broken for political reasons. This may have negatively affected banks’ ability to attract capital and cope with rising asset quality problems. Imposition of a stringent “fit and proper” standard for board members and management and other strict governance rules immediately after the PSI might have improved banks’ governance faster, avoiding the need to police governance problems on a case-by-case basis.26

Figure 19.
Figure 19.

Greece: Banking System Impairment Charges 1/

(Billions of euros)

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: Bank of Greece; European Banking Authority; SNL; and IMF staff calculations.1/ Does not include impairments related to losses on Greek debt holdings under the PSI scheme.2/ Annualized unprovisioned credit losses projected by the Bank of Greece in 2012 capital needs assessment (based on end-2011 data).3/ Annualized unprovisioned credit losses projected by the Bank of Greece in 2014 stress test for the period from end-June 2013 to end-2016 (based on end-June 2013 data).4 /2014 EBA stress test’s projected cumulative impairments under the adverse scenario for 2014-2016 period, annualized (based on end-2013 data).

H. Conditionality

40. Performance with respect to quantitative fiscal targets, especially the quarterly performance criteria (QPCs) was generally strong, in line with other euro area programs (Figure 20). Between March 2012 and December 2013, 79 percent of the QPCs were met, reflecting strong performance on cash primary balance targets. The QPCs on government guarantees were missed in the early phase of the program, while those on arrears were missed frequently. Performance with respect to the ITs was much weaker (15 percent of targets were met), particularly for those on arrears and privatization.27

Figure 20.

EFF Compliance: QPCs

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: IMF MONA database and staff calculations.1/ Calculated as ratio of met to total number of QPCs; excludes criteria that were cancelled.

41. The structural conditionality, however, was much more detailed in Greece vis-à-vis comparable programs, and with lower implementation rates (Figures 21 and 22). The number of structural conditions per review in Greece was much higher than in comparable countries. This reflected two considerations. First, Greece had weaker starting conditions requiring a larger number of conditions to meet ambitious program objectives. Second, the detailed approach to conditionality was meant to ensure substantive implementation of structural measures. The formulation of structural conditionality was broadly consistent with intended structural reforms. However, the prevalence of prior actions, including the conversion of a large number of structural benchmarks into prior actions, frequent implementation delays, and a significant number of non-implemented benchmarks across all structural areas suggest weak ownership. In addition, backtracking on a number of structural measures and nominal implementation of conditionality reflected lack of political consensus in the face of strong opposition of vested interests.

Figure 21.
Figure 21.

Program Conditionality and Performance Conditionality

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: IMF MONA database and staff calculations.1/ Calculated as ratio between number of benchmarks which were met, implemented with delay or partially to the total numbers of benchmarks. This calculation excludes benchmarks that were: a) converted to prior actions; and b) outstanding at the end of the program.
Figure 22.

Greece: Conditionality and Performance by Sector

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Sources: IMF MONA database and staff reports.

I. Collaboration with the Troika Partners

42. The support of Greece’s European partners exceeded the initial program commitments. The EFSF program was fully disbursed despite the EFF going off track. Significant debt relief was provided in the context of the OSI (Box 2). The ECB provided sizeable liquidity support to Greek banks, which experienced deposit outflows. The amount of ECB liquidity support peaked at about €150 billion in 2015 and declined to about €100 billion by the end of the EFF in January 2016. A new ESM program (€86 billion) covering the period of August 2015 to August 2018 was agreed between the EU and Greece in August 2015. The off-track EFF program and the ESM program overlapped during August 2015–January 2016 (with no reviews under the ESM program completed during this period). The dialogue between staff and the Troika partners continued during August 2015–January 2016 but there was a disagreement on key DSA parameters. Compared with staff, the EC had a more optimistic view on the long-term economic outlook, the country’s capacity to sustain primary fiscal surpluses of 3.5 percent of GDP over the medium and long term, and debt sustainability. Greece’s EU partners were also concerned about moral hazard associated with upfront commitments of debt relief.

43. Significant changes in the EMU institutional framework and policies were implemented during the EFF program period. Firewalls were strengthened through the establishment of the ESM in late 2012. A Single Supervisory Mechanism (SSM) was implemented in 2014, and a Single Resolution Mechanism (SRM), which was based on the Bank Recovery and Resolution Directive (BRRD), became operational in 2015. Regarding monetary policy, in 2012, the ECB made it clear that it would do whatever it takes to preserve the euro. Furthermore, with the increasing risks of entrenched deflation and weak activity, the ECB started a quantitative easing program in early 2015. All these measures and policies reduced systemic concerns in the euro area, benefitting Greece as well.

44. The staffs of the three Troika partners were of the view that there is room for improving collaboration. The EC staff felt that the following issues need to be addressed in a more systematic manner: the reconciliation of technical analysis, the division of labor in terms of design and monitoring of conditionality, communication strategies, and information-sharing. The Fund staff raised the issues of sharing confidential information and modalities of assurances regarding euro area-wide policies affecting member countries with Fund-supported programs. These two issues have intensified since the establishment of the SSM in 2014. Staff were not guaranteed sufficient confidential supervisory information on Greece. In addition, in the future, it is not clear how to reconcile potential tensions between the staff’s financial sector advice specific to Greece and relevant EU directives and the EC’s and SSM’s views or to secure EC’s and SSM’s assurances for implementation of agreed measures.

Program Design Issues

45. Drawing on the analysis of the program outcomes, this section raises issues related to the program design strategy. In the case of this EFF, it is difficult to ask hypothetical questions on whether modifying certain program design features at the margin would have delivered materially different outcomes because counterfactual analysis is notoriously speculative in highly volatile crisis situations amidst significant political instability. With the benefit of hindsight, this section focuses on the critical elements of the program design, which in their totality might have helped deliver better outcomes. But better outcomes would not have been guaranteed as political instability and fragile ownership may have doomed any program regardless of its design.

46. Were political economy considerations given sufficient weight in program design? The ambitious reform and adjustment agenda agreed in 2012 required strong ownership and support across the political spectrum. The coalition government that negotiated the EFF made a promising start on reforms. At the same time, the destabilizing political economy implications of the fourth year of the deep recession, falling real incomes, and rising unemployment and poverty, the strong opposition of vested interest to structural reforms, and a significant deterioration in payment culture were recognized as important risks. With the benefit of hindsight, a more gradual pace of fiscal adjustment and a more focused approach to structural reforms within a longer time horizon potentially stood better chances of success. However, it should be recognized that at the time of program discussions, the Greek authorities’ stated commitment to frontloaded fiscal adjustment and a more comprehensive reform agenda, as well as political constraints on official financing, significantly reduced the likelihood of reaching consensus on such an approach quickly.

47. Was the program too optimistic about payoffs of structural reforms? The program implicitly assumed early growth payoffs from structural reforms. Lessons from this approach are threefold. First, it is advisable to use more conservative estimates of the growth benefits of structural reforms, including significant implementation lags.28 Second, there should be a more realistic assessment of the government’s ability to pursue multiple politically difficult reforms simultaneously with the large fiscal consolidation. Third, weak ownership cannot be entirely mitigated through detailed conditionality, regardless of the number of prior actions. Going forward, ownership could be fostered by focusing on a smaller number of reforms with a clear implementation sequencing. Assisting the authorities in devising a supportive communication strategy may also promote ownership.

48. Did delays in financial sector reforms exacerbate output decline? Signs of rising risks to asset quality stemming from previous rapid credit growth and precipitous increases in real estate prices did not receive sufficient attention in 2012. This could be in part explained by the benign growth forecast at the time of the program request, as well as emerging political opposition to removing foreclosure moratoria, political instability, and significant capacity constraints. As a result, many financial sector reforms, including private sector insolvency frameworks and NPL reduction measures, were initiated with a significant delay. The slow pace of balance sheet repair contributed to high NPLs, which in turn created headwinds to credit and real activity.

49. Were programmed fiscal adjustment and the long-term primary surplus targets realistic? Based on panel data for 27 countries, IMF (2015c) argues that fiscal adjustment of more than 5 percent of GDP within three years (or 1.7 percent per year on average) 29 has an increasingly adverse impact on the medium-term debt-to-GDP dynamics. In Greece, this threshold was substantially exceeded during 2010–13 with the cyclically adjusted primary balance improving by 17.3 percent of GDP or 4.3 percent per year on average,30 which most likely contributed to a deterioration of the debt-to-GDP ratios. Also, the pace of Greece’s fiscal adjustment (2010–13) was well above the pre-2008 episodes of large fiscal adjustments in Europe, such as Denmark, Sweden, and Finland, which relied on exchange rate and interest rate instruments to promote export and investment growth.31 It seems that the realized fiscal adjustment in the context of internal devaluation in Greece may have exceeded a “speed limit,” beyond which the economy’s capacity to support a given debt burden is reduced and investor confidence and social cohesion are undermined.32 It should be recognized, however, that determining this speed limit ex ante is very hard, owing to the complexity of interaction between economic and political factors. Regarding the long-term level of the primary surplus, IMF (2016b) provides evidence that in a sample of 55 countries in the last 200 years, there have been only 15 episodes of recessions of longer than 5 years, and no country sustained a primary surplus of larger than 2 percent of GDP after such a period of negative growth.3334 It is questionable whether Greece could set a historic precedent by maintaining primary surpluses in the range of 3.5–4.5 percent of GDP over the medium and long term as assumed under the program—especially in light of the high structural unemployment.

50. Was debt relief sufficient? In light of cross-country evidence (para. 49) and institutional constraints in Greece (para. 46–47), it appears that the program’s objectives for the pace of fiscal adjustment and the long-term primary surplus were in a highly ambitious, virtually unprecedented range with arguably low chances of being achieved. Therefore, a more gradual pace of fiscal adjustment (as staff argued initially) and a lower long-term primary surplus target may have been justified from the outset. This in turn would have required more financing and more upfront debt relief. It is doubtful that additional debt relief in the context of the PSI could have been advisable because a larger haircut could have diluted PSI participation and resulted in greater losses for the domestic banking sector. More debt relief and more financing could have only been provided by official creditors. More debt relief under the OSI was eventually provided in January 2013 and a commitment was made to provide additional debt relief in the future if needed. However, this contingent debt relief was conditional on maintaining the large primary surpluses over the medium and longer terms.

51. Should the Fund insist on the composition of fiscal adjustment? In retrospect, the deviation of the composition of fiscal adjustment from the initial program assumptions reflected political economy and capacity constraints on pursuing higher quality but more difficult reforms. Ultimately, the lower-quality adjustment strategy may have negatively affected growth and social cohesion. Significant increases in regressive taxation (e.g., VAT rate hikes) may have contributed to the decline in political support for the program. Large increases in corporate taxes, with exemptions in place, may have undermined the growth potential of the taxpaying corporations. Also, tax evasion, in particular by the self-employed, contributed to the shift of the burden of adjustment to the poor. In light of this experience, there is merit for the Fund to insist on structural benchmarks and prior actions that would improve the composition of fiscal adjustment while ascertaining that there is strong ownership for the recommended fiscal measures.

52. Are there alternative approaches to assessing debt sustainability in Greece? In retrospect, targeting a specific long-run level of debt as a criterion for debt sustainability might not have been well-founded in Greece’s particular circumstances. Despite uncertainties related to long-term forecasts, the analysis of debt sustainability based on flows (gross financing needs), as well as the trajectory of the stock of debt, as staff proposed in the 2015 DSA, may have been a more suitable framework for Greece since the beginning of the EFF, when long-term concessional official debt replaced private debt.35

53. Staff already began to address these design issues during the EFF, with the work continuing in the course of on-going discussions on a possible follow-up program. As EFF program implementation encountered increasing difficulties under challenging political circumstances, staff was learning from experience and adjusting the program design and policy dialogue accordingly. A more realistic growth and a slower pace of fiscal adjustment were adopted at the time of the first/second reviews. The issues of NPLs and insolvency frameworks started to be addressed from the fifth review and reflected in the staff’s advice after mid-2014. The discussions of the ability of the Greek political system to deliver required reforms quickly, the rationale for a further slowdown in fiscal adjustment, and the re-prioritization of structural reforms gained in prominence after mid-2015. In addition, the EPE on the SBA (2010–12), which was concluded in June 2013 with a delay, covered many of the same issues, and its lessons were progressively incorporated into the EFF program design (Box 1).

Compliance with Fund Policies and Procedures

A. Justification for Exceptional Access

54. According to the Fund policies in force at the time of the program request, any exceptional access arrangement had to satisfy four criteria:

  • Criterion 1—presence of exceptional balance of payment pressures. This criterion was clearly met during the entire program period. While the current account pressures subsided during the program period, the absence of sustained access to capital markets and large capital outflows during the entire duration of the program were prevalent (Figure 2).

  • Criterion 2—debt sustainability with high probability. Staff concluded that debt was sustainable but not with high probability, invoking the systemic exemption36 at the time of the program request and all completed reviews. The 2015 DSA concluded that debt was unsustainable, but by that time the program was off track and so this assessment was made outside the context of an EFF review. The overly optimistic macro forecast and the realization of the very high program implementation risks explain why Greece could not achieve debt sustainability under the EFF program. The systemic exemption was referred to but not justified in detail except at the time of the first/second reviews.37 The systemic exemption claim was finally abandoned, outside the review context, in the standalone 2015 DSA. That DSA recognized that even though the systemic exemption applied in the past, there was “no rationale for continuing to invoke it when debt relief was needed now on the official sector (rather than private) claims.” Had this judgement been reached earlier, it is not clear that a case could have been made for invoking the systemic exemption throughout the EFF, as official debt had largely replaced private claims since the beginning of the program.38

  • Criterion 3—re-accessing capital markets. At the time of the EFF request, Greece had no market access. Program documents argued that market access was expected to be restored only gradually in the post-program period, at short maturity and relatively high interest rates, and conditional on a full implementation of the program. By the time of the fifth review, Greece accessed markets for the first time in four years, taking advantage of the relatively benign macroeconomic situation and the global search for yield. That episode, however, proved to be a fleeting respite, which staff correctly identified as such. In retrospect, the program’s failure to restore debt sustainability and growth, as well as establishment of capital controls, explain why market access was not restored in the post-program period. Nevertheless, given that the judgement underpinning the justification of this criterion at the time was based on the assumption that the program would be implemented, the conclusion that market access could be regained by the end of the program appears reasonable.

  • Criterion 4—the policy program of the member provides reasonably strong prospects of success, including not only the member’s adjustment plans but also its institutional and political capacity to deliver that adjustment. At the time of the program request and in subsequent reviews, staff argued that despite implementation problems, the authorities had demonstrated ownership and policy resolve through the completion of multiple prior actions and were benefitting from significant capacity-improving technical assistance provided by the EC and the Fund (Boxes 36). While willing to give the authorities the benefit of the doubt on this basis, staff nevertheless acknowledged the very high implementation risks, stressing that the program would continue to “test political and social resolve … and the authorities’ administrative capacity.” (IMF 2012, page 40). The staff’s emphasis on the very high program implementation risks against the requirement of “reasonably strong prospects of success” suggests that the bar for meeting Criterion 4 was set too low. In retrospect, despite the large number of prior actions and capacity-building efforts, the implementation risks materialized after each review, undermining the achievement of program objectives.

B. Mix of Adjustment and Financing

55. Available financing required significant fiscal adjustment in the context of optimistic macroeconomic and political economy assumptions. The amount of financing committed to Greece during the EFF period was unprecedented and required arduous political efforts by EU creditor countries: €144.7 billion from the EU (with actual disbursements amounting to €163.2 billion), SDR 23.8 billion from the Fund (of which SDR 10.2 billion disbursed), a haircut on private debt of €106 billion, and ECB liquidity support of up to €150 billion. At the time of the EFF request, the Fund was faced with a stark choice between a program, the virtually unprecedented ambition of which had to match the available financing, and a disorderly Grexit with possibly deep systemic implications against the background of the incomplete architecture of firewalls in Europe. The Fund took a high risk to support the pro-reform coalition government, considering that the alternatives for Greece and the euro area were much worse at the time.

56. Even then, the level of Greece’s access to Fund resources was very high. The proposed access of SDR 23.8 billion (2,159 percent of quota at the time) was the fourth largest in the Fund’s history, after Greece’s SBA, and Ireland’s and Portugal’s EFFs. Despite greater financing needs compared with the SBA, access and phasing under the EFF would be more conservative. The EFF would account for only 16 percent of total (post-PSI/OSI) financing needs, with evenly phased disbursements—17 purchases of SDR 1.4 billion (127 percent of quota) each—to link use of Fund resources to program performance. Nevertheless, under the approved EFF access, the peak access level39 was projected to be the highest in the Fund’s history in terms of quota and the second highest (after Iceland) in terms of GDP. In light of the exceptionally high exposure of the Fund’s balance sheet and program implementation risks, the report for the EFF request could have discussed in more detail the rationale for burden sharing between the Fund and European partners. The use of Fund resources for budget financing purposes was consistent with the Fund’s mandate because Greece had a balance of payment need and fiscal need.

C. Capacity to Repay the Fund

57. The Fund’s exposure to Greece was substantial, and staff was candid about the major risks to the country’s repayment capacity from the beginning.40 Despite the significant PSI and OSI, Greece’s external debt service burden was projected to remain high. With SDR 17.5 billion (of the planned SDR 26.43 billion) purchased under the SBA, Greece’s pre-EFF credit outstanding was already equal to 1,592 percent of quota prior to the EFF. Under the EFF, credit outstanding was projected to peak at 2,570 percent of quota, the highest in the Fund’s history. The program would contribute to the liquidity and concentration risks resulting from the large outstanding amounts of resources used by a few members.

58. Greece accumulated overdue financial obligations to the Fund of SDR 1.6 billion due on June 30 and July 13, 2015, becoming the first advanced country to be in arrears on a payment to the Fund. The missed payments constituted the largest amount of overdue financial obligations from a member country in the Fund’s history. Greece’s non-payments more than doubled the arrears to the Fund’s General Resource Account (GRA) at the time. While the Fund’s balance sheet was sufficiently strong to meet its own financial obligations and potential financing needs by other members, a prolonged period of nonpayment could have had significant implications for the Fund’s finances. If Greece had fallen into protracted arrears (i.e., overdue for more than six months) the carrying value of credit outstanding may have needed to be adjusted, which, would have raised the possibility of the recognition of an impairment loss and a loss of Fund income. Greece made debt service payments to other creditors while having overdue financial obligations to the Fund, which risked undermining the Fund’s preferred creditor status. In the event, Greece’s arrears were short-lived. The Greek government agreed to a new ESM program with additional adjustment measures (para. 42), which helped secure bridge financing to repay the Fund on July 20, 2015.

59. As foreseen by staff, Fund resources were to a large extent safeguarded by the European partners, although the nature of such financing assurances was not well defined. The staff stated that full program implementation and the willingness of the European partners to continue to backstop Greece’s payments capacity to the Fund after the program period (IMF 2012, ¶53) were key to ensuring Greece’s repayment capacity. The EU assurance was in the form of a commitment to “continue providing support to all countries under programs until they have regained market access, provided they fully implement those programs” (Euro Summit Statement, October 26, 2011). The European support, however, was not clearly defined. It was made conditional on Greece fully complying with the requirements and objectives of the program. It was unclear, therefore, what assurances would remain if Greece were to fail to comply with EU-related program objectives after the end of the EFF. Given these considerations, it is evident that such contingent assurances to ensure the capacity to repay the Fund in the future cannot be a substitute for debt sustainability.

Conclusions and Lessons

60. The EFF with Greece posed exceptional challenges to the Fund. After strong pressures from the international community, Greece’s main political parties united behind a technocratic government in late 2011, raising hopes that the mounting implementation problems evident under the 2010 SBA could be overcome. On the strength of this consensus, and reflecting the insistence of Greece’s European partners on adhering to the EU Stability and Growth Pact as soon as possible, the new government proposed an ambitious program. While indicating the advantages of a more gradual fiscal adjustment, staff decided to support the authorities’ ambitious targets in view of the newfound political consensus, a welcome commitment by European partners to stand by Greece, and the still notable systemic risks. However, staff was seriously concerned about downside risks from the outset, and the staff report accompanying the request for the EFF presented the Executive Board with a frank and explicit assessment of such risks, notably those stemming from the ambitious primary surplus targets and the high indebtedness. The Executive Board approved the EFF fully cognizant of the risks. This decision revealed a high risk tolerance by the institution, recognizing that the risks of not continuing to support Greece at the time were greater for Greece and for the euro area.

61. A measure of progress toward the EFF objectives was achieved, but the program ultimately foundered in the face of adverse political developments. During the first two years of the program, despite frequent interruptions, significant fiscal and external adjustment was undertaken, and some structural reforms progressed (e.g., public financial management, elements of pension and labor market reforms, and select financial sector measures). Above all, Greece remained in the euro area, which contributed to a reduction in systemic risks. While the program was launched with broad-based backing from Greece’s main parties, however, political instability subsequently dogged the program, and ultimately derailed it, reflecting fragile ownership and strong opposition from vested interests. In the summer of 2015, Greece became the first advanced economy to accumulate temporary overdue financial obligations to the Fund—the largest in the Fund’s history.

62. Growth, competitiveness, and debt sustainability have not been restored. To achieve these objectives, Greece needs to continue with unfinished reforms and its EU partners need to provide more debt relief.

63. It is possible that any program, no matter how well-designed, could have failed in such difficult circumstances. However, the program’s chances of success might have been somewhat greater if the degree of ambition in its targets and the optimism of its macro assumptions had been tempered. A less ambitious approach would have required more financing and more debt relief from the outset.

64. Staff’s approach evolved as the program risks started to materialize. Staff worked with the authorities and their European partners to revise the initial ambitious program targets, provide additional financing and debt relief, and find ways to re-invigorate stalling reforms. The significant re-design of the program in reviews, and a further recalibration of policy advice once the program was irretrievably off-track, demonstrated that the Fund can learn from experience and adapt its approach to evolving circumstances. It was appropriate for the Fund to interrupt the program when there was growing evidence of insufficient policy commitment or financing to achieve broad program objectives. The program relationship was instrumental for staff’s close engagement with the authorities and their European partners in analyzing policy options and tradeoffs.

65. With the benefit of hindsight, the report derives a number of possible lessons, most of which were applied during the EFF and in the subsequent discussions with the authorities:

  • When the political base for reforms is fragile, program assumptions and design should be more conservative from the start. Political economy constraints, and the impact of the pace and composition of fiscal adjustment, as well as macro-financial linkages, on growth should be better reflected in program design. The staff should resist understandable pressures from the authorities (and in the euro area context, their EU partners) for more optimistic assumptions.

  • Financial sector reforms are essential for economic recovery. Delays in addressing NPLs, private sector insolvency frameworks, and governance issues in the banking sector weighed on the recovery. Steadfast implementation of reforms in these areas should be given high priority.

  • The composition of fiscal adjustment matters for fiscal sustainability and social cohesion. Contrary to the spirit of the program and despite persistent efforts by staff, the composition of fiscal adjustment was not socially equitable, raising concerns about the political sustainability of the achieved fiscal consolidation. Enforcement of tax compliance, development of targeted social safety nets, and pension reform are particularly important for making the adjustment more durable and equitable.

  • Structural reforms require time and strong ownership to bear fruit. Greece needs to restart the stalled reforms, including in the areas of product, service, and labor markets, and regulated professions, to remain a viable euro area member. Securing strong ownership and adopting a more parsimonious approach to structural conditionality are also key. Assumptions regarding the growth payoff from structural reforms need to be conservative.

  • Upfront debt relief commitments consistent with debt sustainability based on a realistic target for the medium-term primary fiscal surplus are a prerequisite for program success in the circumstances faced by Greece. In light of the uncertain nature of conditional assurances of third parties to ensure the capacity to repay the Fund, securing debt sustainability based on realistic assumptions is called for from the outset.

  • There is merit in formalizing the operational framework for Fund collaboration with monetary unions in the program context. A possible agreement should cover the issues of information-sharing, the reconciliation of technical analysis, the division of labor in terms of design and monitoring of conditionality, communication strategies, modalities of assurances regarding union-wide policies affecting program member countries, and financial assurances to be provided to the Fund.

  • Certain Fund policies would benefit from a fresh discussion. In light of the very high risks explicitly documented at the time of the program request and their subsequent realization, including temporary overdue financial obligations to the Fund, there is merit in reviewing Fund risk acceptance guidelines and the exceptional access criterion on prospects for program success.

Conclusions of the EPE on the SBA (2010–12)

The ex post evaluation of the SBA (IMF 2013a), which was concluded in June 2013, identified the following lessons:

  • Better tailoring of Fund lending policies to the circumstances of monetary unions. The report saw merits in an EFF arrangement in light of the structural nature of challenges facing Greece and criticized the baseline macro assumptions as overly optimistic.

  • Avoiding undue delays in debt restructuring. The report argued that earlier debt restructuring could have eased the burden of adjustment and contributed to a less dramatic contraction in output.

  • More attention to the political economy of adjustment. The report emphasized the importance of fighting tax evasion to achieve a more equitable distribution of adjustment costs.

  • More parsimony in fiscal structural reforms. The report underscored that detailed conditionality would not be able to substitute for political ownership.

  • More effective risk-sharing arrangements within the euro area. The report stated that the Greek crisis brought to the fore shortcomings in the euro area related to risk sharing and crisis response.

The 2013 EPE was conducted 15 months after EFF approval. As a result, its conclusions and recommendations could not influence the design of the 2012 EFF. Nevertheless, some lessons of the EPE were accounted for even before its publication: a shift to an EFF in the case of Greece in 2012, debt relief at the outset of the EFF (although it was not sufficient in retrospect), and an intensification of discussions on more effective risk-sharing arrangements within the euro area as part of Fund regional surveillance. Other lessons, including the need for realistic forecasts,1 the importance of political economy factors, more parsimony in structural reform, and the criticality of sufficient debt relief, were only addressed as the EFF implementation encountered significant difficulties. This experience suggests that EPEs should be concluded prior to approval of successor programs even if a compressed production schedule is required.

1 The forecast errors for the EFF at the time of the request were somewhat larger than those for the SBA at the time of the request for the first two program years. The forecast errors declined significantly from the time of the first/second reviews under the EFF.

PSI, OSI, and Debt Sustainability

The Greek debt restructuring of March 2012, the largest in history, was calibrated to bring Greece’s public debt down to 120 percent by 2020. First announced in June 2011, the private sector involvement (PSI) followed months of negotiations among the Greek authorities, Euro group representatives, and the creditor group led by the Institute for International Finance (IIF), with staff participating as observers. The IIF’s PSI proposal that came in July 2011 provided insufficient debt relief and had to be recalibrated, driven partly by the deepening recession.1 Once the terms of the debt exchange were announced in February 2012, however, the PSI took place swiftly. Under the PSI, €197 billion of Greek government bonds (GGBs) were exchanged for €62 billion of new debt and €30 billion in short-term EFSF notes, resulting in a debt write-down of €106 billion or 52 percent of 2012 GDP, a haircut of 53.5 percent in nominal terms.

A01bx2ufig1

Composition of Greek Sovereign Debt 1/

(Billions of euros)

Citation: IMF Staff Country Reports 2017, 044; 10.5089/9781475576252.002.A001

Source: Zettelmeyer, et. al. (2013).1/ Shows Greek government and government-guaranteed debt owed to private and official creditors as of February 2012, i.e. before debt exchange. ECB/NCB debt refers to ECB SMP holdings as well as holdings by national central banks in the Euro area. EU/EFSF loans include the bilateral Greek Loan Facility loans as well as the EFSF loans. T-bills are privately held short-term debt instruments.

Delays in PSI negotiations reduced the stock of debt eligible for the debt exchange. The decision not to restructure debt at the outset of the Greek crisis, grounded in international spillover concerns, had already allowed some €40 billion in maturing bonds to be fully repaid in the first year of the SBA.2 Once the PSI was deemed necessary, the drawn-out negotiations meant that some further €10 billion continued to be repaid in full (Zettelmeyer, Trebesch, and Gulati 2013). By the time of the debt exchange, Greece’s debt was largely held by Greek and European banks, and by the ECB (through Securities Markets Program (SMP)-related bond purchases). The ECB, as well as European national central banks, and the EIB, as official creditors, would be excluded from the PSI.

Several factors ensured the near-universal participation in the PSI. The most important was the retrofitting of “collective action clauses” (CACs) to the outstanding bonds to allow a qualified majority of creditors to legally bind all others to the terms of a debt restructuring. With the vast majority of GGBs issued under local law, this significant contract modification simply required a change in the domestic law. Banks were encouraged to participate in the debt exchange through a combination of moral suasion and official sector pressure, helping to further ensure the activation of CACs. The debt exchange also offered generous near-cash sweeteners in the form of EFSF notes and upgrading the new GGBs to English law bonds to strengthen the incentive for participation. In the end, 97 percent of eligible bonds were tendered for the exchange.

Involving the banking sector in the PSI was inevitable but controversial. Carving out the banking sector from the PSI was not an option given the amount of public debt held by Greek banks. But the PSI further weakened a system that was already impaired by heavy deposit outflows. To contain the fallout on the financial sector, therefore, €50 billion from the program was set aside for bank recapitalization with the capital requirements to be gradually phased in.

At the outset of the program, the envisaged official sector involvement (OSI) was relatively narrow in scope. Interest on the existing Greek Loan Facility (GLF) would be lowered from 300 bps to 150 bps, and national central banks would repatriate the profits on their holdings of Greek bonds back to Greece. New financing through the EFSF would be provided at lower cost and at longer maturities. These concessions, however, would prove inadequate in the ensuing months.

By the first/second EFF reviews, a revised DSA showed that further debt relief would be needed from the European partners. With a significantly worse macroeconomic outlook, the scope of the OSI was expanded to include further lowering of the interest rates on and lengthening of the maturities of GLF and EFSF loans. Around €11 billion in EFSF funding was used upfront to buy back €32 billion of the new GGBs, taking advantage of the low prices. To ensure that the program is financed, the European partners reiterated their commitment to support Greece “as necessary during and beyond the program” contingent upon program implementation.

1 Under the IIF’s original proposal, debt relief for Greece in NPV terms ranged between zero and 11.5 percent, for discount rates between 5 and 9 percent. Using the prevailing “risk free” rate of around 3.5 percent would have resulted in an NPV increase in Greece’s debt (see Zettelmeyer, Trebesch, and Gulati 2013).2 Staff’s estimate, based on Bank of Greece’s gross external debt data and European Commission (2010).

PFM Reforms

PFM reforms under the EFF focused on budgeting and budget monitoring, spending controls, fiscal reporting and the institutional and legal framework. Substantial reforms to PFM systems have been implemented since 2012 with extensive Fund technical assistance:

  • Legal and institutional reforms, including amendments to the Organic Budget Law and a reorganization of functions in the Ministry of Finance, have strengthened responsibility, accountability and processes for effective financial management.

  • Creation of financial management capacity aimed at devolving budget and financial management responsibility through the creation of General Directorates of Financial Services (GDFS) has been legislated and is at an advanced stage of implementation (expected to be completed by January 2017).

  • Monthly published fiscal data covering general government facilitates frequent monitoring of sizeable fiscal activities taking place beyond the State. The coverage of monthly reports is now at a level comparable or better than most advanced countries.

  • Government payment processes have been streamlined and automated to a point where compliance with the requirements of the EU late payment directive is now technically in sight.

  • Cash management operations have been strengthened with the unification of this function in the Public Debt Management Agency.

  • A spending review process has been institutionalized in the Ministry of Finance and pilot reviews have been conducted.

Nevertheless, progress is lacking in a number of areas, such as increasing the staffing and capacity of the GDFS, modernizing payment processes against the resistance of the Hellenic Court of Auditors, the ongoing difficulties to ensure coordination among different stakeholders in fiscal reporting, and the challenges to halt accumulation of new spending arrears.

Reforms of Revenue Administration

At the time of the program request, the reforms were aimed at overhauling tax administration, improving its efficiency and effectiveness, strengthening its enforcement and compliance operations, and enhancing the collection of social security contributions. Extensive Fund TA was provided in support of these objectives.

Greater autonomy of tax administration was a key component of fiscal institutional reforms– including the strengthening of its headquarters, governance, organization, and management—alongside with the revamping of core tax administration operations. Key milestones in revenue administration autonomy were the creation of the General Secretariat for Public Revenues (GSPR) in end-2012 and the first fixed-term (5 years) Secretary General (SG) appointment in January 2013. Substantive reforms were implemented in 2013 and the first half of 2014 due to the strong commitment from the SG in progressing the reform agenda. However, political commitment to respect the fixed-term appointment and provide leadership stability to the revenue administration was absent from mid-2014 to early 2016, when reform progress slowed significantly.

Although some compliance initiatives have been successfully implemented, fundamental weaknesses in core operations remain. Positive initiatives include targeted compliance controls in some taxpayer segments and taxes, and new procedures for monitoring return filing and payment to facilitate immediate action in cases of non-compliance. Tax revenues have stabilized as a percent of GDP, despite the economic contraction that could have increased non-compliance. The most recent results of EC TAXUD’s study of VAT gaps in EU member states (CASE, 2016) shows a decline in Greece’s compliance gap from 36 percent of potential VAT in 2011 to 28 percent in 2014, which is equivalent to a net annual revenue gain of 1.3 percent of GDP. However, these improvements require further consolidation. There are still problems of poor auditing practices and weak recovery of debts that have not been fully modernized yet. Furthermore, ad-hoc schemes, such as amnesties, that may generate short-term revenues but run high moral hazard risks have been counterproductive. As a result, the level of tax arrears continued to grow significantly. Measures to better use Greece’s anti-money laundering framework have generated large numbers of suspicious transaction reports to the Financial Intelligence Unit, but this easily actionable information has received no priority in the tax administration.

Pension Reform

The pension reform focused on long-run sustainability and short-term needs of the pension system. The Fund provided extensive technical assistance on pension reform.

The pension reform started in 2010 with the aim to ensure long-run sustainability of the “main” pensions, equalize rules across pension funds, increase labor force participation, and provide a safety net for the elderly. The pension law of 2010 was scheduled to take effect in 2015, yet it was only partially implemented.

As the fiscal situation deteriorated, pension expenditure reached close to 18 percent of GDP and complexities of the pension system became evident in 2012, the focus shifted from long-term reforms towards measures yielding more immediate savings. In 2012, main pension benefits were reduced by 12 percent above €1,300, and supplementary pension benefits by 10 percent under €250, 15 percent over €250-300, and 20 percent above €300. The supplementary pension system was consolidated. In 2013, the 13th and 14th payments were abolished. Besides benefit cuts, other measures included raising the retirement age by two years, increasing the health care contribution by retirees, and gradual phasing-out of grandfathering rules for retirement. Reversal of reforms occurred with the Council of State’s verdict of the unconstitutionality of progressive pension cuts.

Despite reform measures, the pension system remains highly imbalanced and pension spending stays at 17-18 percent of GDP, up from 14 percent at the onset of the crisis, partly due to the deep recession and deflation. Pension deficits also rose to 11 percent of GDP at end-2015, by far the highest in the Euro Area.

Tax Policy Reforms

The reforms in tax policy focused on revamping tax legislation. The reform of the income tax law (ITL) aimed at the codification of the ‘income tax’-related provisions scattered in the Greek legal system, the simplification of the—often very legalistic—language, closing loopholes by introducing modern anti-avoidance concepts, and broadening the tax base by eliminating personal deductions and streamlining remaining incentives. A new legal framework (Tax Procedure Code—TPC) was needed to enable the modernization of tax administration. The property tax reform began in 2013, with the aim of replacing the outdated wealth tax with a modern property tax system. The objective of the VAT reform was to streamline the system.

With FAD/LEG’s technical assistance, a new income tax law was drafted addressing the issues mentioned in the previous paragraph. In July 2013, lawmakers adopted a new ITC which incorporated many recommended provisions. Over the last three years, successive governments, however, have backtracked substantially on the tax policies that were introduced in the ITC in 2013. The current ITC reversed a number of deductions and incentives, re-introduced complex legalistic language, and contains additional measures—sometimes addressing administrative shortcomings and/or perceived ‘fairness’ issues—that re-introduced new tax planning opportunities.

The TPC, drafted with FAD/LEG’s technical assistance, was passed in July 2013 almost unchanged. Since its passage, the TPC has not been influenced substantially by political interference and still forms a good legal basis for the modernization of the Greek tax administration.

The VAT reform reduced the number of rates and thus broadened the base, but it stopped short of eliminating the intermediate rate and relied on a further hike of the top rate. The special reduced rate applicable on the islands was removed, however.

There was some progress on real estate taxation. The new property tax ENFA was legislated at end-2013, but the reform of the property valuation was stalled due to the lack of political will.

References

  • Alcidi, C., A. Belke, A. Giovannini, and D. Gros. 2016. “Macroeconomic Adjustment Programmes in the Euro Area: An Overall Assessment.International Economics and Economic Policy 13: 345358.

    • Search Google Scholar
    • Export Citation
  • Arghyrou, M., and J. Tsoukalas. 2011. “The Greek Debt Crisis: Likely Causes, Mechanisms and Outcomes.The World Economy, 34 (2): 173191.

    • Search Google Scholar
    • Export Citation
  • Athanasoglou, P., C. Backinezos, and E. Georgiou. 2010. “Export performance, competitiveness and commodity composition.Bank of Greece Working Paper 114.

    • Search Google Scholar
    • Export Citation
  • Blanchard, O. 2015. “Greece: Past Critiques and the Path Forward.iMFdirect, July 9 and July 14.

  • Blanchard, O., and D. Leigh. 2013. “Growth Forecast Errors and Fiscal Multipliers.IMF Working Paper 13/1.

  • Bouis, R., and R. Duval. 2011. “Raising Potential Growth After the Crisis: A Quantitative Assessment of the Potential Gains from Various Structural Reforms in the OECD Area and Beyond,” OECD Economics Department Working Papers 835.

    • Search Google Scholar
    • Export Citation
  • Bouis, R., R. Duval, and J. Eugster. 2016. “Product Market Deregulation and Growth: New Country-Industry-Level Evidence.IMF Working Paper 16/114. Washington.

    • Search Google Scholar
    • Export Citation
  • Bower, U., V. Michou, and C. Ungerer. 2014. “The Puzzle of the Missing Greek Exports.Economic Papers 518, European Commission.

  • Center for Social and Economic Research (CASE). 2015. “Study and Reports on the VAT Gap in the EU-28 Member States: 2016 Final Report.

    • Search Google Scholar
    • Export Citation
  • Cottarelli, Carlo, and Laura Jaramillo. 2012. “Walking Hand-in-Hand: Fiscal Policy and Growth in Advanced Economies.IMF Working Paper N0.12/137 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Dabla-Norris E., G. Ho, K. Kochar, A. Kyobe, and R. Tchaidze. 2014. Anchoring Growth: The Importance of Productivity-Enhancing Reforms in Emerging Market and Developing Economies.

    • Search Google Scholar
    • Export Citation
  • Duval, R., and D. Furceri. 2016. “The Effects of Labor and Product Market Reforms: The Role of Macroeconomic Conditions and Policies.IMF Annual Research Conference, Washington.

    • Search Google Scholar
    • Export Citation
  • Eichengreen, B., and U. Panizza. 2016. “A Surplus of Ambition: Can Europe Rely on Large Primary Surpluses to Solve its Debt Problems?Economic Policy 31, 549.

    • Search Google Scholar
    • Export Citation
  • European Commission. 2010. “The Economic Adjustment Programme for Greece.Occasional Papers 61, May.

  • Gal, P., and A. Hijzen. 2016. “The Short-Term Impact of Product Market Reforms: A Cross-Country Firm-Level Analysis.IMF Working Paper 16/116. Washington.

    • Search Google Scholar
    • Export Citation
  • Gournichas, P., T. Philippon, and D. Vayanos. 2016. “The Analytics of the Greek Crisis.NBER Macroeconomics Annual 31.

  • Higgins, M. and T. Klitgaard. 2011. “Saving Imbalances in the Euro Area Sovereign Debt Crisis.Current Issues in Economics and Finance 17 (5): 111.

    • Search Google Scholar
    • Export Citation
  • Independent Evaluation Office of the International Monetary Fund. 2016. The IMF and the Crises in Greece, Ireland, and Portugal. Washington.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund. 2009. The State of Public Finances Cross-Country Fiscal Monitor: November 2009. Washington.

  • International Monetary Fund. 2010. Ex Post Evaluations of Exceptional Access Arrangements—Revised Guidance Note, Washington.

  • International Monetary Fund. 2012. “Greece: Request for Extended Arrangement Under the Extended Fund Facility.IMF Country Report 12/57.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund. 2013a. “Greece: Ex Post Evaluation of Exceptional Access under the 2010 Stand-By Arrangement.IMF Country Report 13/156.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund. 2013b. “Reassessing the Role and Modalities of Fiscal Policy in Advanced Economies.IMF Policy Paper, September 2013. Washington.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund. 2014. “Adjustment in the Euro Area Deficit Countries: Progress, Challenges, and Policies.Staff Discussion Note 14/7, Washington.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund. 2015a. “Preliminary Draft Debt Sustainability Analysis.IMF Country Report 15/165.

  • International Monetary Fund. 2015b. “An Update of IMF Staff’s Preliminary Public Debt Sustainability Analysis.IMF Country Report 15/186.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund. 2015c. Crisis Program Review. Washington.

  • International Monetary Fund. 2015d. Structural Reforms and Macroeconomic Performance: Country Cases. IMF Policy Papers, November, IMF: Washington DC.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund. 2016a. Fiscal Monitor: October 2016. Washington.

  • International Monetary Fund. 2016b. “Greece: Preliminary Debt Sustainability Analysis: Updated Estimates and Further Considerations.IMF Country Report 16/130.

    • Search Google Scholar
    • Export Citation
  • Kalemli-Oscan, S., L. Laeven and D. Moreno. 2015. Corporate Debt Overhang, Rollover Risk, and Investment in Europe. ECB, CEPR, and University of Maryland.

    • Search Google Scholar
    • Export Citation
  • Le Fouler, L., W. Suyker, and D. Turner. 2001. “Trade Linkages and the Trade Matrices in the OECD Interlink Model”, OECD Economics Department Working Papers 310. Paris.

    • Search Google Scholar
    • Export Citation
  • McKinsey&Company. 2012. Greece 10 Years Ahead: Defining Greece’s New Growth Model and Strategy.

  • OECD. 2012. OECD Economic Outlook, Volume 2012, Issue 1. Paris.

  • Schwab, K. (ed), 2016, The Global Competitiveness Report 2016–2017, World Economic Forum, Geneva.

  • Varga, J., W. Roeger, and J. in’t Veld. 2014. “Growth Effects of Structural Reforms in Southern Europe: The Case of Greece, Italy, Spain, and Portugal.European Commission Economic Papers 511.

    • Search Google Scholar
    • Export Citation
  • World Bank. 2016. Doing Business 2017: Equal Opportunity for All. Washington.

  • Zettelmeyer, J., C. Trebesch, and M. Gulati. 2013. “The Greek Debt Restructuring: An Autopsy,Economic Policy 28, 513563.

Annex I. Authorities’ Views on the EPE Report1

The authorities broadly agreed with most lessons identified in the report. They highlighted that the stakeholders faced the extraordinary challenges of engineering a sizeable fiscal adjustment and internal devaluation simultaneously in a relatively closed economy belonging to a monetary union.

The authorities agreed with the general principle that the program should have a realistic macroeconomic framework and targets from the outset. They saw the initial program macroeconomic framework as excessively optimistic and the program targets as too ambitious. They argued that program underperformance is mainly explained by underestimation of fiscal multipliers, the lack in the initial program design of an appropriate sequencing of structural reforms, and the underestimation of the implementation challenges and negative political repercussions of recommended policies (both of which ultimately undermined ownership). However, the former authorities in charge of the initial program discussions indicated that there was no alternative to the frontloaded adjustment in light of political constraints on available financing and debt relief and the need to strengthen credibility; and their preference for the ambitious program should be understood in this context. At the same time, the Bank of Greece representatives indicated that, at times (e.g., in 2013), the staff was excessively pessimistic regarding expected yields of fiscal measures, which resulted in budget over-performance, leading to an excessive fiscal tightening.

The authorities disagreed with the report’s analysis of the composition of fiscal adjustment during the program. While the ministry of finance acknowledged that previous governments agreed that fiscal adjustment should have been implemented mainly on the expenditure side and through permanent measures, the current position of the government is that most of the adjustment should have occurred on account of revenue measures aimed at reducing tax avoidance and tax evasion, as well as governance improvements in tax administration. The government believes that there has been a large compliance gap, the elimination of which should be given the utmost priority. The expenditure-to-GDP ratio in Greece is broadly consistent with the EU averages and does not require further adjustments, in their view. In addition, the Bank of Greece representatives argued that the composition of fiscal adjustment should be supplemented by an analysis of revenue and expenditure in terms of nominal values rather than solely based on the changes in the ratios of total revenue and total expenditures to shrinking GDP. According to the Bank of Greece, the country achieved a very sizeable decline in nominal primary expenditure (excluding recapitalization costs, Figure 12b) in large part through nominal reductions in wages and pensions. The Bank of Greece is also of the view that the program should have focused on structurally adjusted fiscal targets, but acknowledged that this approach would have been constrained by lack of additional financing.

The authorities broadly agreed with the lessons from implementation of the financial sector measures under the program. They felt that the mounting problems with the insolvency frameworks and rising NPLs required earlier attention. Furthermore, the Bank of Greece representatives stated that the incidents raising potential questions with respect to banks governance are currently being addressed based on the new regulations that are consistent with the report’s recommendations. Finally, the Bank of Greece maintained that the report’s assessment that initial recapitalizations should have relied on more conservative assumptions is only possible with the benefit of hindsight, as the stress tests had to rely on official program forecasts. While acknowledging that three rounds of recapitalization turned out to be necessary during the program period, the Bank of Greece representatives underlined that the total need for public funds had been over-estimated by the staff by about €10 billion.

The authorities argued that the program had failed to incorporate the appropriate sequencing of structural reforms and to take into account their political feasibility and the capacity constraints facing the government. For example, some product market reforms, with an uncertain impact on alleviating the economy-wide supply bottlenecks or benefitting consumers, generated strong political backlash, thereby eroding the successive governments’ political capital. Also, the large number of reforms overwhelmed the available administrative capacity. Furthermore, the Bank of Greece stated that the report downplayed progress in structural reforms. In their view, implemented labor market and other reforms contributed to reducing ULCs and improving cost competitiveness, which explains the nascent recovery in exports, excluding shipping. Nevertheless, the authorities agreed that there was a need for further progress in improving price and non-price competitiveness, while more work needed to be done to tailor the necessary reforms to Greece’s circumstances. Regarding the EU context for structural reforms, the authorities felt that there was a need for stronger progress on positive integration (i.e., promoting union-wide institutions) before intensifying negative integration (i.e., reducing barriers).

The authorities agreed that sufficient upfront debt relief and adequate financing are important prerequisites for program success. They also stated that a piecemeal approach to providing debt relief is counterproductive. Furthermore, the Bank of Greece representatives argued that there was a missed opportunity to request higher OSI in the first half of 2014.

The authorities agreed with the recommendation that there is merit in formalizing the operational framework for Fund collaboration with monetary unions. In their view, the frequent disagreements among the Troika partners complicated program discussions and implementation.

Table A1.1.

Greece: Selected Economic Indicators 1/

article image
Sources: Bank of Greece; Eurostat; Hellenic Statistical Authority; Ministry of Economy and Finance; and IMF staff estimates.

Actual data for 2010 and 2011 differ from the EFF Request data due to revisions by Hellenic Statistical Authority. Fiscal data starting in 2012 onwards are based on EFF program definition.

Based on the Labor Force Survey.

Table A1.2.

Greece: Quantitative Performance Criteria and Indicative Targets (2012–13)

(Billions of Euros, unless otherwise indicated)

article image
Sources: IMF MONA database and staff reports.

For September 2013, the target on privatization was set as an indicative target (IT) at the time of the EFF request in the amount of €1.2 billion. At the 1st and 2nd Review, its status was changed to a quantitative performance criterion (QPC) with a target of €1.8 billion (but only for the September 2013 test date, while remaining an IT for December 2013 and afterwards). This September 2013 QPC was subsequently revised (while remaining as a QPC) to €0.9 billion at the 4th Review. At the same time, the IT for December 2013 was revised to €1.6 billion (down from the initial target of €2.5 billion, as shown in the table). At the 5th Review, for the September 2013 test date, this measure was assessed as an IT against a target of €1.5 billion.