## 3.1  The Spillover Index - Vector Autoregressive Model

## Source details

**Canonical URL:** [3.1  The Spillover Index - Vector Autoregressive Model](https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022133-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2022/english/wpiea2022133-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2022/english/wpiea2022133-print-pdf.pdf.json)

---

### Overview
- Context: Collaboration between the IMF and European lending mechanisms during the European debt crisis involved differing priorities: the IMF’s focus on "the stability of the international monetary and financial system" versus the EC and ECB emphasis on reducing contagion within the euro area.
- Research aim: Assess how the presence of contagion (spillovers) within a region affects the design of lending programs offered by regional mechanisms, and whether spillover risk gives borrowing countries bargaining power that yields more favorable borrowing conditions.
- Approach:
  - Novel dataset of announcements and disbursements (April 29, 2009 until September 20, 2018) for crisis-hit euro area countries (Ireland, Greece, Spain, Portugal, Cyprus).
  - Compute measures of interconnectedness using a VAR-based spillover index (Diebold and Ylmaz, 2009) and Dynamic Conditional Correlations from a bivariate-GARCH DCC model (Engle, 2002a).
  - Develop a recursive contract model with spillover costs (limited commitment principal-agent).
  - Validate model predictions empirically using disbursement-based proxies for current consumption.

### Dataset on loan announcements and disbursements
- Coverage: Ireland, Greece, Spain, Portugal, Cyprus.
- Sample period: April 29, 2009 until September 20, 2018 (database records daily announcements; final sample covers 2007 to 2018 in descriptive facts).
- For each recorded day the dataset includes:
  - whether a statement relating to a lending program was issued by the EC/ESM or the IMF;
  - whether an EC/ESM statement mentions a concrete future or current-period loan disbursement;
  - whether a realized EC/ESM disbursement was anticipated;
  - the time horizon at which a future EC/ESM disbursement is anticipated (within the current month or the next 1, 2, 3, 6, 12, 24 or 36 months);
  - the announced or realized loan size, interest rate and maturity of the loan.
- Data fields: daily dummy variables for items 1–3 (equal one if information provided on a given day, zero otherwise), plus loan characteristic fields (size, interest rate, maturity).
- Novelty: First dataset to record all dates of official meetings and statements by EC/ESM/IMF for these programme countries and to distinguish announcements of anticipated vs. realized disbursements and revisions.

### Empirical measures of contagion and spillovers
- Methods:
  - VAR-based aggregate spillover index (Diebold and Ylmaz, 2009) computed in subsamples before and after the peak of the crisis.
  - Time-varying correlations estimated from a bivariate-GARCH Dynamic Conditional Correlation model (DCC) (Engle, 2002a) used as a proxy for contagion/interconnectedness.
- Interpretation: Contagion treated as an increasing function of spillovers; higher spillovers imply higher cost to partner countries and greater contagion risk.

### Main empirical findings — VAR, DCC, and descriptive facts
- VAR/DCC findings:
  - Evidence of contagion between euro-area countries, but not full transmission (spillover measures < 100 percent), implying room for risk-sharing.
  - Heterogeneous interlinkages: peripheral countries (Italy, Spain, Portugal, Ireland) are major sources of spillovers among themselves and affect some core countries; Cyprus contributes least to variability of sovereign and banking markets of other countries.
  - DCC early-sample (2009) correlations for Greece, Ireland, Portugal with euro-area markets: 0.6-0.7.
  - 2012 regime switch to a "low-contagion" regime as investors diversified away.
  - Spillover index figures (selected):
    - Generalized Δ(Sovereign CDS), full-sample: Spillover Index 54.09%.
    - 10 yr Government Bonds Total Returns: Spillover Index 76.17%.
    - Equity Total Returns: Spillover Index 79.38%.
    - Generalized Δ(Bank CDS): Spillover Index 58.77%.
    - Subsample 2008-2011 Generalized Δ(Sovereign CDS): Spillover Index 73.63% (reported elsewhere as 76.63% for 2008-2011 in text summary—both figures preserved as presented in source).
- Descriptive facts and timing:
  - Greece and Ireland were the most common subject of statements and press releases by the EC and the ESM followed by Spain, Portugal and Cyprus.
  - Large gap between "all" announcements and concrete references to future/current disbursements, especially for Spain.
  - For Greece, future disbursements were announced three times more often than actual disbursements; for Ireland the ratio was below 1.
  - Key crisis lending events (verbatim chronology preserved):
    - April 2010: Greek authorities requested an initial loan of 45 billion.
    - May 2010: announced a number of austerity measures to secure a three-year 110 billion loan.
    - February 2012: a second bailout loan of 130 billion was approved.
    - 2013: the Greek parliament approved new austerity measures paving the way for a new tranche worth almost 7 billion euros.
    - 2015: the Troika approved the third bailout package worth 86 billion euro.
    - Portugal received a 78 billion bailout package but regained complete access to financial markets in September 2013.
    - On 9 June 2012, the Eurogroup decided to grant Spain a financial support package with a maximum value ofe100 billion.
    - Cyprus requested assistance in 2012; the package worthe10 billion was approved in March/April 2013.

### Spillovers and announcements — regression evidence
- Strategy: Regress DCC on a dummy equal to one on days of announcements (announcements by European lending mechanisms and by the IMF), controlling for Euribor and Eurostoxx50 returns.
- Main regression findings:
  - Announcements by European lending mechanisms are negatively related to spillovers for Ireland, Greece and Portugal; positive but not significant for Cyprus in pooled regressions.
  - Lending-related announcements by the IMF have no significant effect on spillovers for debtor countries in the presented specifications.
- Country-level coefficients from Appendix F (Table 11 and Table 12 preserved):
  - Regressing DCC on EFSM/EFSF/ESM (EC) announcements (Observations 2452 each):
    - IRL_DCC: EC announcements coefficient −0.0338 ∗∗∗ (t = −3.49); Constant 0.321 ∗∗∗ (t = 193.28).
    - GRE_DCC: EC announcements coefficient −0.0381 ∗∗ (t = −2.99); Constant 0.109 ∗∗∗ (t = 48.19).
    - PRT_DCC: EC announcements coefficient −0.0430 ∗∗∗ (t = −3.48); Constant 0.308 ∗∗∗ (t = 166.38).
    - CYP_DCC: EC announcements coefficient 0.00697 (t = 1.40); Constant 0.0922 ∗∗∗ (t = 151.52).
  - Regressing DCC on IMF announcements (Observations 2452 each):
    - IRL_DCC: IMF announcements −0.0151 (t = −1.47); Constant 0.321 ∗∗∗ (t = 192.52).
    - GRE_DCC: IMF announcements −0.0000469 (t = −0.00); Constant 0.108 ∗∗∗ (t = 47.85).
    - PRT_DCC: IMF announcements −0.0185 (t = −1.70); Constant 0.308 ∗∗∗ (t = 165.04).
    - CYP_DCC: IMF announcements −0.00451 (t = −1.09); Constant 0.0923 ∗∗∗ (t = 151.64).
- Disaggregated European announcements (future disbursements vs current-period disbursement):
  - Announcements of future loan disbursements and of current-period disbursement significantly reduce spillovers for Ireland and Portugal; coefficients not significant for Greece and Cyprus in the disaggregated specifications (Tables 13–16).

### Official loan terms (selected figures reported in the source)
- Greece:
  - May 2010: EFSF/ESM Maturity 5 yrs Interest Rate 4.04; IMF Maturity 15 yrs Interest Rate 3.23
  - June 2011: EFSF/ESM Maturity 10 yrs Interest Rate 3.78; IMF Maturity 5 yrs Interest Rate 3.53
  - March 2012: EFSF/ESM Maturity 20 yrs Interest Rate 2.07; IMF Maturity 8 yrs Interest Rate 3.13
  - December 2012: EFSF/ESM Maturity 30 yrs Interest Rate 0.93; IMF Maturity 8 yrs Interest Rate 3.07
- Ireland:
  - December 2010: EFSF/ESM Maturity 7.5 yrs Interest Rate 5.25; IMF Maturity 7 yrs Interest Rate 3.37
  - July 2011: EFSF/ESM Maturity 15 yrs Interest Rate 2.74; IMF Maturity 7 yrs Interest Rate 3.53
  - June 2013: EFSF/ESM Maturity 22 yrs Interest Rate 2.32; IMF Maturity 8 yrs Interest Rate 3.07
- Portugal:
  - May 2011: EFSF/ESM Maturity 7.5 yrs Interest Rate 5.47; IMF Maturity 7 yrs Interest Rate 3.37
  - July 2011: EFSF/ESM Maturity 15 yrs Interest Rate 3.15; IMF Maturity 7 yrs Interest Rate 3.53
  - June 2013: EFSF/ESM Maturity 22 yrs Interest Rate 2.19; IMF Maturity 8 yrs Interest Rate 3.07
- Spain:
  - November 2012: EFSF/ESM Maturity 12.5 yrs Interest Rate 0.78; IMF terms not listed
- Cyprus:
  - May 2013: EFSF/ESM Maturity 15 yrs Interest Rate 1.03; IMF Maturity 4 yrs Interest Rate 3.07

### Theoretical model and comparative-static predictions
- Model type: Recursive dynamic principal-agent model with one-sided limited commitment (borrower can renege to autarky); lender fully committed.
- Novel element: Introduces spillover costs ζ(w_j,ˆs′) that the principal incurs whenever the borrower credibly threatens to default (binding participation constraint); ζ_w < 0, ζ_ww > 0 (ζ strictly decreasing and convex in promised utility).
- Key constraints (notation preserved):
  - Participation constraint (PC): u(c_t) + β w_t ≥ u(y_t) + β v_aut ∀t.
  - Promise keeping constraint (PKC): Σ Π_j (u(c_j)+β w_j) ≥ v.
  - State variable ˆs = max{y_0,...,y_t} captures when PC binds in future states.
- First-order conditions under binding PC (preserved forms):
  - u′(c_j) = − Π_j θ / (Π_j θ + λ_j) + ... (as given in source equations).
  - P′(w_j) − ζ′(w_j) = −(θ + λ_j Π_j).
- Comparative-static prediction (Proposition 2 preserved):
  - For the same past promised utility v, the contract with present spillover cost allocates higher promised utility and lower current consumption compared to the contract without spillover cost:
    - w_spil_j > w_no spil_j
    - c_spil_j < c_no spil_j
- Mechanism intuition:
  - Spillover costs make the lender vulnerable to borrower default; to reduce expected spillover cost the lender back-loads promised transfers (increasing future promised utility w_j and lowering current consumption c_j), because ζ is decreasing in w_j making future promises relatively less costly at margin.

### Empirical validation of model predictions — current consumption and disbursements
- Measure of current consumption: daily ratio = cumulative disbursements / total loan amount promised by European lending mechanisms (EFSF, EFSM, ESM, GLF) from press releases (EC and ESM).
- Descriptive: for Greece, Portugal, Ireland, Cyprus cumulative disbursements fell short of initial promises except for Ireland in some periods.
- Panel regression specification (equation (3) preserved):
  - Current Consumption_it = c + β DCC_MA_it + u_it
  - DCC_MA: past moving averages of DCC over 30 and 90 days before t; also standard DCC measures used.
- Regression results (Appendix F, Table 10 preserved; Observations 7716 each, fixed effects and clustered standard errors):
  - Model (1): DCC coefficient −2.049** (t = −7.61); Constant 1.227*** (t = 19.89); Observations 7716.
  - Model (2): MA DCC[-30] coefficient −2.136** (t = −6.36); Constant 1.249*** (t = 16.16); Observations 7716.
  - Model (3): MA DCC[-90] coefficient −2.242* (t = −5.06); Constant 1.278** (t = 12.41); Observations 7716.
- Interpretation: Higher interconnectedness (DCC measures) is negatively related to current consumption (cumulative disbursements), consistent with model prediction that contracts with high spillovers back-load transfers.
- Robustness: Regression coefficients remain negative and significant at the 5% and 10% level respectively for all DCC measures used; results robust to linear de-trending.

### Country-specific lending program facts and disbursement timing (selected excerpts)
- Portugal timeline highlights (preserved chronological entries):
  - 1.19 May 2011: "Various borrowing operations by EFSM and EFSF will take place between 23 May and 15 July to cover first disbursements to Portugal and Ireland for a total of 15.3 billions".
  - 2.12 August 2011: "Approval of the conclusion of this review will allow the disbursement of 11.5 billion (7.6 billion by the EU, and 3.9 billion by the IMF). This disbursement can take place in September".
  - 3.16 November 2011: "Approval of the conclusion of this review will allow the disbursement of 8 billion (5.3 billion by the EU, and 2.7 billion by the IMF). These disbursements could take place in December and January.".
  - 4.28 February 2012: "Approval of the conclusion of this review will allow the disbursement of 14.9 billion (9.7 billion by the EU, and 5.2 billion by the IMF). These disbursements could take place in April subject to the approval of the IMF Executive Board and ECOFIN and EUROGROUP".
  - Subsequent reviews and disbursements through 12.2 May 2014 with noted shortfalls and reconciliations.
- Table 17 excerpt — Loans disbursed to Portugal (selected rows, exact values preserved):
  - (01) 31/05/2011 — Disbursement (bn euros): 1.75 — Cumulative Disbursement (bn euros): 1.75
  - (02) 01/06/2011 — Disbursement (bn euros): 1.50 — Cumulative Disbursement (bn euros): 3.25
  - (03) 01/06/2011 — Disbursement (bn euros): 2.25 — Cumulative Disbursement (bn euros): 5.50
  - (04) 01/06/2011 — Disbursement (bn euros): 1.00 — Cumulative Disbursement (bn euros): 6.50
  - (05) 22/06/2011 — Disbursement (bn euros): 6.50 — Cumulative Disbursement (bn euros): 3.70
  - ...
  - (28) 12/11/2014 — Disbursement (bn euros): 0.40 — Cumulative Disbursement (bn euros): 24.30 — Cumulative Total: 26.00
- Discrepancies noted for Portugal:
  - EFSM agreed to disburse 26 bn in total but disbursed 24.3 bn (missing 1.7bn).
  - Specific review shortfalls: 3rd Review 0.1 bn short; 4th Review 0.1 bn short; 6th Review 0.8 bn short; 11th Review 0.4 bn short; 12th Review 1.7 bn never disbursed.

### Conclusion (preserved statements)
- Presence of contagion/spillover costs influenced lending contract design within the euro-area: contagion reduced current consumption for crisis countries and back-loaded future consumption.
- Newly built dataset records dates of official meetings and announcements by IMF and European institutions and includes rich loan-condition and disbursement timing information.
- Empirical evidence: European Mechanisms’ decisions to provide financial support successfully mitigated spillovers for several programme countries; IMF announcements did not show a significant spillover-reducing effect in the presented specifications. European lending programs with high spillover considerations tend to promise more future utility and deliver less current-period disbursements, consistent with the theoretical recursive-contract predictions.

*Source: wpiea2022133-print-pdf - 3.1  The Spillover Index - Vector Autoregressive Model.*

### 3.1  The Spillover Index - Vector Autoregressive Model. . . . . . . . . . .   11

### 3.1  The Spillover Index - Vector Autoregressive Model

### Overview
- Context: Collaboration between the IMF and European lending mechanisms during the European debt crisis involved differing priorities: the IMF’s focus on "the stability of the international monetary and financial system" versus the EC and ECB emphasis on reducing contagion within the euro area.
- Research aim: Assess how the presence of contagion (spillovers) within a region affects the design of lending programs offered by regional mechanisms, and whether spillover risk gives borrowing countries bargaining power that yields more favorable borrowing conditions.
- Approach: Build a novel dataset of announcements and disbursements (April 29, 2009 until September 20, 2018) for crisis-hit euro area countries (Ireland, Greece, Spain, Portugal, Cyprus); compute measures of interconnectedness using a VAR-based spillover index and dynamic conditional correlations from a bivariate-GARCH model; develop a recursive contract model with spillover costs; validate model predictions empirically using disbursement-based proxies for current consumption.

### Dataset on loan announcements and disbursements
- Coverage: All euro area crisis countries: Ireland, Greece, Spain, Portugal, Cyprus.
- Sample period: April 29, 2009 until September 20, 2018.
- For each day recorded, the dataset includes:
  - whether a statement relating to a lending program was issued by the EC/ESM or the IMF;
  - whether an EC/ESM statement mentions a concrete future or current-period loan disbursement;
  - whether a realized EC/ESM disbursement was anticipated;
  - the time horizon at which a future EC/ESM disbursement is anticipated to take place (within the current month or the next 1, 2, 3, 6, 12, 24 or 36 months);
  - the announced or realized loan size, interest rate and maturity of the loan.
- Data fields: daily dummy variables for items 1–3 (equal one if information provided on a given day, zero otherwise), plus loan characteristic fields (size, interest rate, maturity).
- Novelty: First dataset to record all dates of official meetings and statements by EC/ESM/IMF for these programme countries and to distinguish announcements of anticipated vs. realized disbursements and revisions.

### Empirical measures of contagion and spillovers
- Methods used:
  - VAR-based aggregate spillover index (Diebold and Ylmaz, 2009) computed in subsamples before and after the peak of the crisis.
  - Time-varying correlations estimated from a bivariate-GARCH Dynamic Conditional Correlation model (Engle, 2002a) used as a proxy for contagion/interconnectedness.
- Interpretation: Contagion is treated as an increasing function of spillovers; higher spillovers imply higher cost to partner countries and greater contagion risk.

### Main empirical findings
- Effect of announcements:
  - Lending-related announcements by the European mechanisms reduced the spillover effect that Ireland, Greece and Portugal had on the rest of the union.
  - Lending-related announcements by the European mechanisms increased spillovers for Cyprus.
  - Lending-related announcements by the IMF did not significantly affect spillovers of debtor countries.
- Interpretation: European lending mechanisms appear to have successfully pursued the objective of reducing contagion within the euro area for several programme countries; heterogeneity across countries (e.g., Cyprus vs Ireland) reflects differing cross-country creditor ownership and consequent perceived spillover externalities.

### Official loan terms (selected figures reported in the source)
- Greece:
  - May 2010: EFSF/ESM Maturity 5 yrs Interest Rate 4.04; IMF Maturity 15 yrs Interest Rate 3.23
  - June 2011: EFSF/ESM Maturity 10 yrs Interest Rate 3.78; IMF Maturity 5 yrs Interest Rate 3.53
  - March 2012: EFSF/ESM Maturity 20 yrs Interest Rate 2.07; IMF Maturity 8 yrs Interest Rate 3.13
  - December 2012: EFSF/ESM Maturity 30 yrs Interest Rate 0.93; IMF Maturity 8 yrs Interest Rate 3.07
- Ireland:
  - December 2010: EFSF/ESM Maturity 7.5 yrs Interest Rate 5.25; IMF Maturity 7 yrs Interest Rate 3.37
  - July 2011: EFSF/ESM Maturity 15 yrs Interest Rate 2.74; IMF Maturity 7 yrs Interest Rate 3.53
  - June 2013: EFSF/ESM Maturity 22 yrs Interest Rate 2.32; IMF Maturity 8 yrs Interest Rate 3.07
- Portugal:
  - May 2011: EFSF/ESM Maturity 7.5 yrs Interest Rate 5.47; IMF Maturity 7 yrs Interest Rate 3.37
  - July 2011: EFSF/ESM Maturity 15 yrs Interest Rate 3.15; IMF Maturity 7 yrs Interest Rate 3.53
  - June 2013: EFSF/ESM Maturity 22 yrs Interest Rate 2.19; IMF Maturity 8 yrs Interest Rate 3.07
- Spain:
  - November 2012: EFSF/ESM Maturity 12.5 yrs Interest Rate 0.78; IMF terms not listed
- Cyprus:
  - May 2013: EFSF/ESM Maturity 15 yrs Interest Rate 1.03; IMF Maturity 4 yrs Interest Rate 3.07

### Theoretical model and predictions
- Model type: Recursive contract (principal-agent, constrained efficient allocation) with limited commitment; principal is the lending mechanism, agent is the borrowing country.
- Novel element: Introduces spillover costs that the principal incurs whenever the borrower credibly threatens to default (binding participation constraint). The principal internalizes spillover costs weighted by the probability of a binding participation constraint.
- Comparative static prediction relative to benchmark without spillovers:
  - High spillovers lead the principal to put more weight on future payments compared to the benchmark model without spillovers.
  - The principal substitutes current consumption transfers with future ones to equate the binding participation constraint at lower expected spillover cost, resulting in stronger back-loading of consumption (promised transfers allocated more to the future).
- Contract interpretation: Financial assistance programs act as a carrot-and-stick; voluntary participation constraint is modeled as the credible threat of default/autarky.

### Empirical validation of model predictions
- Proxy for current consumption: current cumulative loan disbursements.
- Empirical result: Higher interconnectedness between the troubled country and the rest of the euro-area partners has a negative effect on the country’s current consumption (i.e., countries with higher spillovers receive less current-period disbursements), consistent with the model prediction that contracts with high spillovers back-load transfers.
- Welfare implication from model and evidence: Borrowing country worse off in the short term but better off in the long term under lending contracts involving high spillovers.

### Additional empirical and methodological notes
- The dataset records anticipated disbursement horizons with granularity (current month, next 1, 2, 3, 6, 12, 24, 36 months) and distinguishes anticipated, realized, and revised dates of disbursements.
- The paper connects empirical VAR-spillover measures and dynamic correlations to the theoretical limited-commitment contract framework to interpret how contagion considerations shape loan design.

*Source: wpiea2022133-print-pdf - 3.1  The Spillover Index - Vector Autoregressive Model. . . . . . . . . . .   11*

### 2013. In contrast, our database contains higher frequency (daily) data on lending

### wpiea2022133-print-pdf - 2013. In contrast, our database contains higher frequency (daily) data on lending related announcements.

### Data and descriptive facts
- Database: daily announcements by the EC and the IMF on lending-related statements and press releases (records statements on the next Monday when issued during the weekend).
- Sample period: final sample covers the period between 2007 to 2018.
- Crisis-country event counts (descriptive facts from Table 4 and Figure 7):
  - Greece and Ireland were the most common subject of statements and press releases by the EC and the ESM followed by Spain, Portugal and Cyprus.
  - Large gap between "all" announcements and concrete references to future/current disbursements, especially for Spain (reflecting Spanish hesitation to ask for support).
  - For Greece, future disbursements were announced three times more often than actual disbursements; for Ireland the ratio was below 1.
- Key crisis lending events (textual chronology preserved exactly):
  - April 2010: Greek authorities requested an initial loan of 45 billion.
  - May 2010: announced a number of austerity measures to secure a three-year 110 billion loan.
  - February 2012: a second bailout loan of 130 billion was approved.
  - 2013: the Greek parliament approved new austerity measures paving the way for a new tranche worth almost 7 billion euros.
  - 2015: the Troika approved the third bailout package worth 86 billion euro.
  - Portugal received a 78 billion bailout package but regained complete access to financial markets in September 2013.
  - On 9 June 2012, the Eurogroup decided to grant Spain a financial support package with a maximum value ofe100 billion.
  - Cyprus requested assistance in 2012; the package worthe10 billion was approved in March/April 2013.
- Observed timing pattern:
  - IMF and EC statements tended to move in parallel, but most initial IMF statements occurred earlier.
  - Actual disbursements were always pre-announced, producing likely anticipation effects.

### Empirical analysis — contagion measures and financial data
- Objectives:
  - Construct an overall measure of contagion (Spillover Index) based on a VAR following Diebold and Ylmaz (2009).
  - Estimate time-varying correlations using a Bivariate-GARCH Dynamic Conditional Correlation Model (DCC) (Engle, 2002a/b).
- Financial data used:
  - Sovereign CDS (5-year seniority) daily and monthly, and 10-year government bond total returns denominated in euro (Datastream).
  - Bank channel: 5-year senior bank CDS daily for largest banks per country (Datastream, euro), equity total return indices from MSCI Bloomberg, consolidated banking data from BIS.
  - Weighted aggregate bank CDS index per country (weights depend on size of asset holdings of each bank; data on size from S&P Global Market Intelligence).
  - Controls: Eurostoxx 50 index, 3-month Euribor, EONIA rate.
  - Daily sovereign CDS analysis for Belgium, Cyprus, France, Germany, Greece, Ireland, Italy, Portugal, Spain (y_t = (y_BEL,t, y_CYPR,t, y_FRA,t, y_DEU,t, y_GRE,t, y_IRL,t, y_ITA,t, y_PRT,t, y_ESP,t)).
  - VAR specification: two lags for endogenous variables; daily data on sovereign CDS (2008-2018).
- Findings from VAR-based Spillover Index:
  - Evidence of contagion between euro-area countries, but not full transmission (spillover measures < 100 percent), implying room for risk-sharing.
  - Heterogeneous interlinkages: peripheral countries (Italy, Spain, Portugal, Ireland) are major sources of spillovers among themselves and affect some core countries.
  - Cyprus does not contribute materially to variability of sovereign and banking markets of other countries.
- DCC time-varying correlations:
  - Early-sample (2009) correlations for Greece, Ireland, Portugal with euro-area markets: 0.6-0.7.
  - 2012 regime switch to a "low-contagion" regime as investors diversified away.
  - For the high-contagion sub-sample (2008-2011) re-run VAR results:
    - Spillover index = 76.63 percent for 2008-2011.
    - Spillover index = 54.09 percent for the entire sample.
    - Spain contributed the most to variation in many countries (notably Italy, Portugal, Ireland); Ireland and Greece also had high contributions; Cyprus the smallest contribution.

### Spillovers and lending — regression evidence
- Strategy:
  - Regress DCC on a dummy equal to one on days of announcements (announcements by European lending mechanisms and by the IMF).
- Main regression findings:
  - Announcements by European lending mechanisms are negatively related to spillovers for all countries except Cyprus (Table 11).
  - Lending-related announcements by the IMF have no significant effect on spillovers for any debtor countries (Table 12).
  - Disaggregated European announcements (future disbursements vs current-period disbursement), controlling for Eurostoxx50 returns and Euribor:
    - Announcements of future loan disbursements and of current-period disbursement significantly reduce spillovers for Ireland (Table 13) and Portugal (Table 14).
    - Coefficients not significant for Greece (Table 15) and Cyprus (Table 16).
  - Empirical implication: European Mechanisms’ announcements achieved the objective of reducing spillovers; IMF announcements did not show a significant spillover-reducing effect.

### Theoretical model — recursive contract with spillover costs
- Framework:
  - Recursive dynamic principal-agent model with one-sided limited commitment (borrower can renege to autarky); lender fully committed.
  - Borrower receives iid endowment stream {y_t}; participation constraint (PC): u(c_t) + β w_t ≥ u(y_t) + β v_aut ∀t.
  - Promise keeping constraint (PKC): Σ Π_j (u(c_j)+β w_j) ≥ v.
  - Additional state variable ˆs = max{y_0,...,y_t} to capture when PC binds in future states.
- Lender profit with spillover costs:
  - Profit value function P(v,ˆs) with expected current transfer Σ Π_j {y_j − c_j} plus discounted expected profits.
  - When PC binds in future states, lender incurs spillover cost ζ(w_j,ˆs′) with ζ_w < 0, ζ_ww > 0 (strictly decreasing and convex in promised utility).
- First-order conditions under binding PC (Proposition 1; notation preserved):
  - u′(c_j) = − Π_j θ / (Π_j + λ_j)  + λ_j term structure as given in the source (equations 1 and 2 preserved in text).
  - P′(w_j) − ζ′(w_j) = −(θ + λ_j Π_j).
- Comparative statics (with vs without spillovers):
  - First-order conditions table reproduced (no Spillover vs Spillover).
  - Proposition 2 (preserved statement): For the same past promised utility v, the contract with present spillover cost allocates higher promised utility and lower current consumption compared to the contract with absent spillover cost.
    - w_spil_j > w_no spil_j
    - c_spil_j < c_no spil_j
- Mechanism intuition:
  - Spillover costs make lender vulnerable to borrower default; to reduce the probability of default and associated costs the lender "back-loads" the allocation — lowers current consumption and increases promised utility (future consumption).
  - Increasing promised utility is less costly due to ζ being decreasing in w_j, so lender relaxes PC via higher future promised utility rather than higher current consumption.

### Empirical validation of model predictions
- Measure of current consumption:
  - Construct daily ratio = cumulative disbursements / total loan amount promised by European lending mechanisms (EFSF, EFSM, ESM, GLF) from press releases (EC and ESM).
  - Denominator largely unchanged for Cyprus, Ireland, Portugal; for Greece denominator changed across GLS, EFSM, ESM programs.
  - Figure 8 (textual description): for Greece, Portugal, Ireland, Cyprus, cumulative disbursements fell short of initial promises except for Ireland.
- Panel regression specification (preserved):
  - Current Consumption_it = c + β DCC_MA_it + u_it  (equation (3)).
  - DCC_MA: past moving averages of DCC over 30 and 90 days before t; also standard DCC measures used.
- Regression results (summary from Appendix E and main text):
  - Higher spillovers (DCC measures) are negatively related to current consumption.
  - Regression coefficients remain negative and significant at the 5% and 10% level respectively for all DCC measures used.
  - Results robust to linear de-trending.
  - Conclusion: higher spillover costs imply lower current disbursement (back-loading) and thus higher consumption allocated to future transfers.

### Conclusion (preserved statements)
- Presence of contagion/spillover costs influenced lending contract design within the euro-area: contagion reduced current consumption for crisis countries and back-loaded future consumption.
- Newly built dataset records dates of official meetings and announcements by IMF and European institutions and includes rich loan-condition and disbursement timing information.
- Empirical evidence: European Mechanisms’ decisions to provide financial support successfully mitigated spillovers; spillover costs are relevant for contract design.

*Italic: Source: wpiea2022133-print-pdf (excerpt provided).*

### References

### References

### Bibliographic sources
- Abrahám, A., E. Cárceles Poveda, Y. Liu, and R. Marimon (2018). On the optimal design of a financial stability fund.
- Alvarez, F. and U. J. Jermann (2000). Efficiency, equilibrium, and asset pricing with risk of default. Econometrica 68(4), 775–797.
- Atkeson, A. (1991). International lending with moral hazard and risk of repudiation. Econometrica: Journal of the Econometric Society, 1069–1089.
- Beirne, J. and M. Fratzscher (2013). The pricing of sovereign risk and contagion during the european sovereign debt crisis. Journal of International Money and Finance 34, 60–82.
- Constancio, V. (2012). Contagion and the european debt crisis. Financial Stability Review 16, 109–121.
- Corsetti, G., A. Erce, and T. Uy (2017). Official Sector Lending Strategies During the Euro Area Crisis. ADEMU Working paper series No. 070.
- Darvas, Z. (2017). Regional and global financial safety nets : the recent European experience and its implications for regional cooperation in Asia. Bruegel working paper No. 06.
- Diebold, F. X. and K. Ylmaz (2009). Measuring Financial Asset Return and Volatility Spillovers, with Application to Global Equity Markets. The Economic Journal 119(1), 158–171.
- Dovis, A. (2019). Efficient sovereign default. The Review of Economic Studies 86(1), 282–312.
- Eaton, J. and M. Gersovitz (1981). Debt with potential repudiation: Theoretical and empirical analysis. The Review of Economic Studies 48(2), 289–309.
- Eichengreen, B. (2015). The Irish Crisis and the EU from a Distance. IMF seminar presentation January 20, 1–14.
- Engle, R. (2002a). Dynamic conditional correlation: A simple class of multivariate generalized autoregressive conditional heteroskedasticity models. Journal of Business & Economic Statistics 20(3), 339–350.
- Engle, R. (2002b). Dynamic Conditional Correlation Dynamic Conditional Correlation : A Simple Class of Multivariate Generalized Autoregressive Conditional Heteroskedasticity Models. No. 0015(2002).
- Forbes, K. J. (2012). The big c: identifying and mitigating contagion. In Proceedings—Economic Policy Symposium—Jackson Hole, pp. 23–87. Federal Reserve Bank of Kansas City.
- Gourinchas, P. O., P. Martin, and T. Messer (2018). The economics of sovereign debt, bailouts and the eurozone crisis. In West Coast Workshop International Finance & Open Economy Macroeconomics. Working Paper, University of California, Berkeley, Berkeley, CA.
- IMF (2017). Collaboration between Regional Financiang arrangements and the IMF. IMF Background paper July 2015.
- Kehoe, P. J. and F. Perri (2002). International business cycles with endogenous incomplete markets. Econometrica 70(3), 907–928.
- Ljungqvist, L. and T. J. Sargent (2018). Recursive macroeconomic theory. MIT press.
- Muller, A., K. Storesletten, and F. Zilibotti (2019). Sovereign debt and structural reforms. American Economic Review 109(12), 4220–59.
- Thomas, J. and T. Worrall (1994). Foreign direct investment and the risk of expropriation. The Review of Economic Studies 61(1), 81–108.

### Appendix A — Financial linkages across the euro area (illustrative findings)
- Data source: BIS Consolidated banking statistics; indebtedness of banks in Cyprus, Ireland, Spain, Portugal and Greece expressed as shares of creditors’ GDP in year of first financial assistance program.
- Main creditor countries identified: Germany, France, the Netherlands and Belgium.
- Notable bilateral exposures:
  - Germany held Spanish banking debt worth 49% of German GDP.
  - France held Greek banking debt worth 28% of French GDP.
  - Portugal exposed to Spanish banking debt equal to 10.4% of Portuguese GDP.
  - Spain held Portuguese banking debt worth 5.6% of Spanish GDP.
  - Greece held Cypriot debt worth 4.8% of Greek GDP.
- Banking sectors with largest debt abroad (relative to their own GDP): Ireland 124% of its GDP; Cyprus 99% of its GDP.
- Measured as a share of creditor countries’ GDP, Cypriot banks posed the lowest threat among the five crisis countries (partly due to the small size of the Cypriot economy).
- Figure references (no files listed): Figures 1–6 illustrate outstanding claims on Irish, Cypriot, Greek, Spanish, and Portuguese banks and share of general government gross debt held by foreign investors (Data sources: BIS Consolidated banking statistics; IMF Sovereign Investor Base Dataset for Advanced Economies).

### Theoretical Appendix — Key analytical results and propositions
- Lagrangian setup distinguishes non-state contingent multiplier θ (PKC) and state-contingent λ_j (PC). The Lagrangian includes terms:
  - Σ_{j=1}^S Π_j[{y_j−c_j}+βP(w_j)] − β Σ_{j=hat{s}}^S Π_j ζ(w_j) + θ[Σ_{j=1}^S Π_j[u(c_j)+βw_j]−v] + Σ_{j=1}^S λ_j[u(c_j)+βw_j−u(y_j)−βv_aut].
- Two cases analyzed for first-order conditions:
  - Case 1: Non-binding participation constraint (PC), λ_j = 0 for j∈[1,max{s}_t^0−1]:
    - u′(c_j) = −1/θ
    - P′(w_j) = −θ
    - Implication: contract attains first best allocation; w_j = v and c_j constant.
  - Case 2: Binding PC, λ_j > 0 for j∈[max{s}_t^0,S]:
    - u′(c_j) = − Π_j θ / (Π_j θ + λ_j)
    - P′(w_j) − ζ′(w_j) = − (θ + λ_j/Π_j)
    - Combined: u′(c_j) = [P′(w_j) − ζ′(w_j)]^{−1}
- Envelope condition: P′(v) = −θ.
- Proposition 2 comparative statics (spillovers vs. no spillovers):
  - Define P′(w_spil_j) − ζ′(w_spil_j) = P′(v) − λ_spil_j Π_s (Spillovers case).
  - Define P′(w_no spil_j) = P′(v) − λ_no spil_j Π_s (No spillovers case).
  - Given λ_spil_j < λ_no spil_j, derive P′(w_spil_j) − ζ′(w_spil_j) > P′(w_no spil_j).
  - With P′′()<0, deduce w_spil_j > w_no spil_j, implying presence of spillover costs leads to increased promised utility.
  - Consumption comparison: u′(c_no spil_j) < u′(c_spil_j) and since u′′()<0 implies c_no spil_j > c_spil_j (i.e., consumption lower when spillovers present).

### Appendix C — Descriptive statistics from database on loan announcements and disbursements
- Example statements (verbatim excerpts preserved):
  - 03/05/2010: Euro area and IMF agreement on financial support programme for Greece: euro area Member States pledged a three-year programme total of 80 billion in bilateral loans; up to e30 billion available for 2010; IMF stand-by arrangement of aboute30 billion, bringing joint commitment to e110 billion.
  - 14/09/2011: Commission proposes better financial terms for EU loans to Ireland and Portugal: reduce margins of 292.5 bps for Ireland and 215 bps for Portugal to zero; extend maturity of individual future tranches from maximum of 15 years to up to 30 years; average maturity from 7.5 years to up to 12.5 years.
  - 03/06/2013: Statement on third review of Spain’s financial assistance programme: programme remains on track; sovereign and corporate bond yields dropping; liquidity situation of Spanish banking sector improved; next review foreseen September 2013.
- Descriptive counts (notes): dataset of EC and IMF announcements between April 29, 2009 and September 20, 2018 shows more numerous IMF announcements due to broader transparency and communication strategy (press releases, public information notices, mission statements).
- Table excerpt (counts by country and announcement type):
  - Columns include counts of days with statements by EC and IMF and disaggregation into "all", "future loan", and "disbursement".
  - Example numeric entries shown: CYP 28 13 10 41; ESP 49 42 n.A.; GRE 60 15 55 57; IRL 57 18 21 51; PRT 46 12 10 55 (preserved as presented).

### Appendix D — Empirical analysis: VAR analysis, spillover indices, and tables
- Spillover indices (selected highlights, exact table values preserved):
  - Table 5 (Generalized Δ(Sovereign CDS), full-sample spillover index): Spillover Index 54.09%.
    - Contribution To Others totals: BEL 107.07, CYP 8.31, FRA 110.93, DEU 48.59, GRE 29.87, IRL 143.52, ITA 202.02, PRT 168.89, ESP 191.07.
    - Contribution From Others row totals include BEL 91.40, CYP 0.98, FRA 18.02, DEU 4.85, GRE 2.72, IRL 16.58, ITA 24.23, PRT 16.92, ESP 20.15, Total 104.45.
  - Table 6 (10 yr Government Bonds Total Returns): Spillover Index 76.17%.
    - Contribution To Others totals: AUT 386.62, BEL 351.98, FIN 131.02, FRA 381.10, DEU 345.31, GRE 35.39, IRL 145.60, ITA 232.09, NLD 373.25, ESP 263.83.
    - Contribution From Others totals: AUT 397.79, BEL 352.24, FIN 91.47, FRA 389.15, DEU 340.25, GRE 39.75, IRL 157.73, ITA 247.04, NLD 366.48, ESP 265.07.
  - Table 7 (Equity Total Returns): Spillover Index 79.38%.
    - Contribution To Others totals: AUT 321.71, BEL 339.92, FIN 250.39, FRA 472.72, DEU 424.95, GRE 209.61, IRL 244.24, ITA 383.39, NLD 430.01, ESP 369.92.
    - Contribution From Others totals: AUT 332.61, BEL 354.68, FIN 214.34, FRA 474.19, DEU 402.39, GRE 220.97, IRL 246.11, ITA 390.45, NLD 433.99, ESP 377.11.
  - Table 8 (Generalized Δ(Bank CDS)): Spillover Index 58.77%.
    - Contribution To Others totals: DEU 223.58, FRA 225.34, GRE 14.31, BEL 77.24, ITA 190.18, PRT 129.69, IRL 40.85, NLD 206.87.
    - Contribution From Others totals: DEU 220.73, FRA 221.41, GRE 17.21, BEL 74.13, ITA 181.53, PRT 141.66, IRL 43.28, NLD 208.11.
  - Table 9 (Subsample 2008-2011 Generalized Δ(Sovereign CDS)): Spillover Index 73.63%.
    - Contribution To Others totals: BEL 308, CYP 22, FRA 264, DEU 221, GRE 249, IRL 271, ITA 332, PRT 336, ESP 357.
    - Contribution From Others totals: BEL 95.60, CYP 4.018, FRA 43.04, DEU 35.27, GRE 31.54, IRL 37.2, ITA 53.58, PRT 45.28, ESP 53.86, Total 304 (as presented).

### Appendix E — Spillovers and loan conditions
- Figure 8 (illustration): Spillover measures and current consumption by country. Current consumption ratio defined as ratio of cumulative loan amounts disbursed over total amounts promised by European lending mechanisms.

### Appendix F — Regression results: Effect of spillovers on current consumption and announcements on DCC
- Regression of current consumption (Cum.loans/Overall) on DCC measures (Table 10):
  - Model (1): DCC coefficient −2.049** (t = −7.61); Constant 1.227*** (t = 19.89); Observations 7716.
  - Model (2): MA DCC[-30] coefficient −2.136** (t = −6.36); Constant 1.249*** (t = 16.16); Observations 7716.
  - Model (3): MA DCC[-90] coefficient −2.242* (t = −5.06); Constant 1.278** (t = 12.41); Observations 7716.
  - Notes: "DCC" is current dynamic conditional correlation of sovereign CDS spreads; panel countries: Greece, Ireland, Portugal, Cyprus; fixed effects and clustered standard errors at country level.
- Regressing DCC on EFSM/EFSF/ESM (EC) announcements (Table 11):
  - IRL_DCC: EC announcements coefficient −0.0338 ∗∗∗ (t = −3.49); Constant 0.321 ∗∗∗ (t = 193.28); Observations 2452.
  - GRE_DCC: EC announcements coefficient −0.0381 ∗∗ (t = −2.99); Constant 0.109 ∗∗∗ (t = 48.19); Observations 2452.
  - PRT_DCC: EC announcements coefficient −0.0430 ∗∗∗ (t = −3.48); Constant 0.308 ∗∗∗ (t = 166.38); Observations 2452.
  - CYP_DCC: EC announcements coefficient 0.00697 (t = 1.40); Constant 0.0922 ∗∗∗ (t = 151.52); Observations 2452.
  - Controls: Euribor, Eurostoxx50 return.
- Regressing DCC on IMF announcements (Table 12):
  - IRL_DCC: IMF announcements −0.0151 (t = −1.47); Constant 0.321 ∗∗∗ (t = 192.52); Observations 2452.
  - GRE_DCC: IMF announcements −0.0000469 (t = −0.00); Constant 0.108 ∗∗∗ (t = 47.85); Observations 2452.
  - PRT_DCC: IMF announcements −0.0185 (t = −1.70); Constant 0.308 ∗∗∗ (t = 165.04); Observations 2452.
  - CYP_DCC: IMF announcements −0.00451 (t = −1.09); Constant 0.0923 ∗∗∗ (t = 151.64); Observations 2452.
  - Controls: Euribor, Eurostoxx50 return.
- Interpretation: EC announcements are associated with statistically significant reductions in DCC for Ireland, Greece, and Portugal in the presented specifications; IMF announcements show smaller and generally statistically insignificant coefficients in the country-level panel regressions.

### Appendix G — Country-specific regressions and notes
- Ireland (Table 13):
  - All EFSM/EFSF/ESM announcements: IRL_DCC coefficient −0.0338 ∗∗∗ (t = −3.49).
  - EFSM/EFSF/ESM loan announcements: −0.0392 ∗ (t = −2.28).
  - EFSM/EFSF/ESM loan disbursements: −0.0441 ∗∗ (t = −2.78).
  - IMF announcements: −0.0151 (t = −1.47).
  - Euribor positive and significant: 0.131 ∗∗∗.
  - Observations: 2452 per specification.
- Portugal (Table 14):
  - All EFSM/EFSF/ESM announcements: PRT_DCC −0.0430 ∗∗∗ (t = −3.48).
  - EFSM/EFSF/ESM loan announcements: −0.0503 ∗ (t = −2.17).
  - EFSM/EFSF/ESM loan disbursements: −0.0622 ∗ (t = −2.56).
  - IMF announcements: −0.0185 (t = −1.70).
  - Euribor positive and significant: 0.135 ∗∗∗.
  - Observations: 2452 (one model 2451).
- Greece (Table 15):
  - All EFSM/EFSF/ESM announcements: GRE_DCC −0.0381 ∗∗ (t = −2.99).
  - EFSM/EFSF/ESM loan announcements: −0.00708 (t = −0.28).
  - EFSM/EFSF/ESM loan disbursements: 0.0377 (t = 0.86).
  - IMF announcements: −0.0000469 (t = −0.00).
  - Euribor strongly positive and significant: 0.218 ∗∗∗.
  - Observations: 2452.
- Cyprus (Table 16):
  - All EFSM/EFSF/ESM announcements: CYP_DCC 0.00697 (t = 1.40).
  - EFSM/EFSF/ESM loan announcements: −0.00641 (t = −0.88).
  - EFSM/EFSF/ESM loan disbursements: −0.00159 (t = −0.19).
  - IMF announcements: −0.00451 (t = −1.09).
  - Euribor positive and significant (small magnitude): 0.00345 ∗∗∗.
  - Observations: 2452.
- Notes on the Portuguese lending programme (G.1):
  - Portugal received financial assistance during 2011-2014.
  - Programme supported by loans from the European Union amounting to 52 bne and a 26 bne fr in the Extended Fund Facility with the IMF (text preserved as in source).
  - European mechanisms’ support equally divided to the EFSM and the EFSF, each responsible for disbursing 26 bne.
  - EFSM did not disburse the total amount scheduled; it was 1.7 bne less than planned.
  - Programme outline published in June 2011 under title "The Economic Adjustment for Portugal".
  - Monitoring via 11 reviews; statements released in each review outlined scheduled disbursements and timelines; comparisons made between outlined future payments and realised disbursements (referenced table 12 in source).

*Content unit: wpiea2022133-print-pdf - References*

### 1.19 May 2011:EU and EFSF funding plans to provide financial assistance for

### 1.19 May 2011:EU and EFSF funding plans to provide financial assistance for Portugal and Ireland

### Timeline of announced reviews and disbursements for Portugal
- 1.19 May 2011: "Various borrowing operations by EFSM and EFSF will take place between 23 May and 15 July to cover first disbursements to Portugal and Ireland for a total of 15.3 billions".
  - In the aforementioned period 6 disbursements took place [(1)-(6)], Four of which from the EFSM and the remaining two from the EFSF. (1)-(4) amount for 6.5 bn (EFSM) and (5)-(6) from EFSF amounting for 5.9 bn.
- 2.12 August 2011: "Approval of the conclusion of this review will allow the disbursement of 11.5 billion (7.6 billion by the EU, and 3.9 billion by the IMF). This disbursement can take place in September".
  - In the aforementioned period the only disbursements were done by the EFSF and correspond exactly to the amount of money announced (7.6 bn) these are the disbursements (7)-(9). The EFSF did not do any payments for this Page 50 / 56 period.
- 3.16 November 2011: "Approval of the conclusion of this review will allow the disbursement of 8 billion (5.3 billion by the EU, and 2.7 billion by the IMF). These disbursements could take place in December and January.".
  - In the aforementioned period, EFSM made one payment (12) while EFSF made three payments (10)-(11) and (13). All those together amount for 5.2 billions (0.1 billion less than announced).
- 4.28 February 2012: "Approval of the conclusion of this review will allow the disbursement of 14.9 billion (9.7 billion by the EU, and 5.2 billion by the IMF). These disbursements could take place in April subject to the approval of the IMF Executive Board and ECOFIN and EUROGROUP".
  - In the aforementioned period, EFSM made 2 disbursements (14)-(15) amounting for 4.5 billions, and EFSF made another two (16)-(17) amounting for 5.2 billions. In total 9.7 billions as announced.
- 5.04 June 2012: "Approval of the conclusion of this review will allow the disbursement of 4.1 billion (2.7 billion by the EU, and 1.4 billion by the IMF). These disbursements could take place in July".
  - In the aforementioned period, only the EFSF made two disbursements, amounting for 2.6 billions (0.1 bn less than announced). (18)-(19) disbursements.
- 6.11 September 2012: "Approval of the conclusion of this review will allow the disbursement of 4.3 billion (2.8 billion by the EU, and 1.5 billion by the IMF). These disbursements could take place in October subject to the approval of the IMF Executive Board and ECOFIN and EUROGROUP.".
  - In the aforementioned period, two disbursements were made. One from the EFSM (20) of 2 billions, and one from EFSF (21) of 0.6 billions amounting in total of the announced amount.
- 7.19 November 2012: "Approval of the conclusion of this review will allow the disbursement of 2.5 billion (1.6 billion by the EU, and 0.9 billion by the IMF). These disbursements could take place in January 2013".
  - In the aforementioned period, January 2013 no disbursement was made by either of EFSM, EFSF. The only disbursement close to this period was 0.6 bn by EFSF in February 2013 (22) in the table. (This is 0.8 bn less than what was promised).
- 8.15 March 2013: "The conclusion of this review could take place in May, subject to the approval of ECOFIN and EUROGROUP and of the IMF Executive Board, and will allow the disbursement of 2.0 billion (1.3 billion by the EU, and about 0.7 billion by the IMF)".
  - In the aforementioned period only EFSF made two disbursements (23) and (24) amounting to 2.1 bn euros (Note here that this amount is more than the announced.  HOWEVER, from previous reviews we have 0.1 bn not disbursed from what was promised in the 2nd Review, 0.1 bn not disbursed from the 4th review and 0.8 bn not disbursed from the 5th review. Adding those disbursements to the 1.3 promised bn sums up to the 2.1 bn disbursed in the aforementioned period).
- 9.3 October 2013: "The conclusion of the 8th and 9th reviews could take place in November, subject to the approval of ECOFIN and Eurogroup and of the IMF Executive Board. This would allow for the disbursement of 5.6 billion (3.7 billion by the EU, and about 1.9 billion by the IMF) following the approval of the current reviews."
  - In the aforementioned period the only payment was done by EFSF and amounts to 3.7 bn (25)
- 10.16 December 2013: "Approval of the conclusions of this review will allow the disbursement of EUR 2.7 billion (EUR 1.8 billion by the EU and EUR 0.9 billion by the IMF), bringing the total amount disbursed to Portugal to EUR 74 billion representing roughly 94 percent of total available financial assistance".
  - Only EFSM made one disbursement (26) of exactly 1.8 bn as announced.
- 11.28 February 2014: "The conclusion of the 11th review could take place in April 2014, subject to the approval of ECOFIN and Eurogroup and of the IMF Executive Board.  This would allow for the disbursement of 2.5 billion (1.6 billion by the EU, and about 0.9 billion by the IMF) following the approval of the current review."
  - In the aforementioned period, EFSF made a disbursement (27) amounting to 1.2 bn euros (0.4 bn less than promised).
- 12.2 May 2014: "The conclusion of the 12th review could take place in June, subject to the approval of the ECOFIN Council and of the IMF Executive Board. This would allow for the disbursement of 2.6 billion (1.7 billion by the EU, and about 0.9 billion by the IMF) following the approval of the current review."
  - Note that at this point EFSM and EFSF owes to Portugal 0.4 bn euros from the 11th review plus another 1.7 bn euros promised by the 12th review. However, the last payment disbursed takes place on 12 November 2014 by the EFSM and amounts to 0.4 bn euros (probably what was owed by the 11th review. The 1.7 bn euros from the 12th review were never disbursed).
  - This can be confirmed by the fact that in total the EFSM had agreed to disburse in total 26 bn euros throughout the programme, however, it only did disburse 24.3 bn (less than agreed by the missing 1.7bn disbursement).

### Table 17 excerpt — Loans disbursed to Portugal (selected rows, exact values preserved)
- (01) 31/05/2011 — Disbursement (bn euros): 1.75 — Cumulative Disbursement (bn euros): 1.75 — Cumulative Total: 1.75
- (02) 01/06/2011 — Disbursement (bn euros): 1.50 — Cumulative Disbursement (bn euros): 3.25 — Cumulative Total: 3.25
- (03) 01/06/2011 — Disbursement (bn euros): 2.25 — Cumulative Disbursement (bn euros): 5.50 — Cumulative Total: 5.50
- (04) 01/06/2011 — Disbursement (bn euros): 1.00 — Cumulative Disbursement (bn euros): 6.50 — Cumulative Total: 6.50
- (05) 22/06/2011 — Disbursement (bn euros): 6.50 — Cumulative Disbursement (bn euros): 3.70 — Cumulative Total: 10.20
- (06) 29/06/2011 — Disbursement (bn euros): 6.50 — Cumulative Disbursement (bn euros): 2.20 — Cumulative Total: 5.90
- (07) 21/09/2011 — Disbursement (bn euros): 5.00 — Cumulative Disbursement (bn euros): 11.50 — Cumulative Total: 5.90
- (08) 29/09/2011 — Disbursement (bn euros): 2.00 — Cumulative Disbursement (bn euros): 13.50 — Cumulative Total: 5.90
- (09) 06/10/2011 — Disbursement (bn euros): 0.60 — Cumulative Disbursement (bn euros): 14.10 — Cumulative Total: 5.90
- (10) 20/12/2011 — Disbursement (bn euros): 14.10 — Cumulative Disbursement (bn euros): 1.00 — Cumulative Total: 6.90
- (11) 12/01/2012 — Disbursement (bn euros): 14.10 — Cumulative Disbursement (bn euros): 1.70 — Cumulative Total: 8.60
- (12) 16/01/2012 — Disbursement (bn euros): 1.50 — Cumulative Disbursement (bn euros): 15.60 — Cumulative Total: 8.60
- (13) 19/01/2012 — Disbursement (bn euros): 15.60 — Cumulative Disbursement (bn euros): 1.00 — Cumulative Total: 9.60
- (14) 24/04/2012 — Disbursement (bn euros): 1.80 — Cumulative Disbursement (bn euros): 17.40 — Cumulative Total: 9.60
- (15) 04/05/2012 — Disbursement (bn euros): 2.70 — Cumulative Disbursement (bn euros): 20.10 — Cumulative Total: 9.60
- (16) 30/05/2012 — Disbursement (bn euros): 20.10 — Cumulative Disbursement (bn euros): 3.50 — Cumulative Total: 13.10
- (17) 30/05/2012 — Disbursement (bn euros): 20.10 — Cumulative Disbursement (bn euros): 1.70 — Cumulative Total: 14.80
- (18) 17/07/2012 — Disbursement (bn euros): 20.10 — Cumulative Disbursement (bn euros): 1.50 — Cumulative Total: 16.30
- (19) 17/07/2012 — Disbursement (bn euros): 20.10 — Cumulative Disbursement (bn euros): 1.10 — Cumulative Total: 17.40
- (20) 30/10/2012 — Disbursement (bn euros): 2.00 — Cumulative Disbursement (bn euros): 22.10 — Cumulative Total: 17.40
- (21) 03/12/2012 — Disbursement (bn euros): 22.10 — Cumulative Disbursement (bn euros): 0.80 — Cumulative Total: 18.20
- (22) 07/02/2013 — Disbursement (bn euros): 22.10 — Cumulative Disbursement (bn euros): 0.80 — Cumulative Total: 19.00
- (23) 27/06/2013 — Disbursement (bn euros): 22.10 — Cumulative Disbursement (bn euros): 1.05 — Cumulative Total: 20.05
- (24) 27/06/2013 — Disbursement (bn euros): 22.10 — Cumulative Disbursement (bn euros): 1.05 — Cumulative Total: 21.10
- (25) 22/11/2013 — Disbursement (bn euros): 22.10 — Cumulative Disbursement (bn euros): 3.70 — Cumulative Total: 24.80
- (26) 25/03/2014 — Disbursement (bn euros): 1.80 — Cumulative Disbursement (bn euros): 23.90 — Cumulative Total: 24.80
- (27) 28/04/2014 — Disbursement (bn euros): 23.90 — Cumulative Disbursement (bn euros): 1.20 — Cumulative Total: 26.00
- (28) 12/11/2014 — Disbursement (bn euros): 0.40 — Cumulative Disbursement (bn euros): 24.30 — Cumulative Total: 26.00

### Discrepancies, shortfalls, and reconciliations (Portugal)
- Multiple reviews show promised EU disbursements that were not fully matched by EFSM/EFSF:
  - 3rd Review: announced EU share 5.3 billion; actual combined disbursements for that period amounted to 5.2 billions (0.1 billion short).
  - 4th Review: announced EU share 2.7 billion; actual EFSF disbursements in period amounted to 2.6 billions (0.1 bn short).
  - 6th Review: announced EU share 1.6 billion; no disbursement in January 2013; only 0.6 bn by EFSF in February 2013 (0.8 bn short).
  - 11th Review: announced EU share 1.6 billion; EFSF disbursed 1.2 bn (0.4 bn short).
  - 12th Review: announced EU share 1.7 billion; 1.7 bn were never disbursed (EFSM agreed to disburse 26 bn total but disbursed 24.3 bn — missing 1.7bn).
- Cumulative reconciliation note: over- and under-disbursements in different reviews may offset (example under/over amounts summed in the 7th review note to explain a 2.1 bn disbursement).

### Notes on the Greek lending programme (G.2)
- Total disbursed history high: 283.7 bn (financial assistance provided by European mechanisms to Greece across programmes and enhanced surveillance).
- 1st programme (Greek Loan Facility - GLF):
  - Began May 2010 and ended June 2013.
  - Disbursed 80 billionse 20 in total and consisted of billateral loans pooled by the European Commission. (Footnote: "The amount was subsequently reduced by 2.7 billionesince Slovakia refused to participate and Portugal and Ireland were expempted after entering a lending programme themselves.")
  - IMF disbursed in total 30 billionethrough a Stand By Arrangement (SBA).
  - Table 18 (GLF disbursements):
    - (1) 18 May 2010 — Disbursement (bn euros): 14.50 — Cumulative: 14.50
    - (2) 13 September 2010 — Disbursement (bn euros): 6.50 — Cumulative: 21.00
    - (3) 19 January 2011 — Disbursement (bn euros): 6.50 — Cumulative: 27.50
    - (4) 16 March 2011 — Disbursement (bn euros): 10.90 — Cumulative: 38.40
    - (5) 15 July 2011 — Disbursement (bn euros): 8.70 — Cumulative: 47.10
    - (6) 14 December 2011 — Disbursement (bn euros): 5.80 — Cumulative: 52.90
- 2nd programme (EFSF):
  - Fully operational since August 2010.
  - Started March 2012 and ended June 2015.
  - In total disbursed 141.8 billionewith loan repayments scheduled from 2023 to 2070 and a weighted average maturity of 42.45 years.
  - Committed all unreleased payments remaining from the GLF (11.8 bne) plus 130 billione.
  - Table 19 (selected sub-tranches and cumulative totals preserved as presented):
    - (1) March/June 2012 (74 bn) — 12 March/10April/25April 2012: Disbursement (bn euros): 29.7 — Cumulative: 29.7; 12 March/10April/25April 2012: Disbursement (bn euros): 4.9 — Cumulative: 34.6; 19 March 2012: Disbursement (bn euros): 5.9 — Cumulative: 40.5; 10 April 2012: Disbursement (bn euros): 3.3 — Cumulative: 43.8; 19 April 2012: Disbursement (bn euros): 25.0 — Cumulative: 68.8; 10 May 2012: Disbursement (bn euros): 4.2 — Cumulative: 73.0; 28 June 2012: Disbursement (bn euros): 1.0 — Cumulative: 74.0
    - (2) December 2012/May 2013 (49.1 bn) — 17/19 December 2013: Disbursement (bn euros): 34.3 — Cumulative: 108.3; 28/31 January 2013: Disbursement (bn euros): 9.2 — Cumulative: 117.5; 28 February 2013: Disbursement (bn euros): 2.8 — Cumulative: 120.3; 3 May 2013: Disbursement (bn euros): 2.8 — Cumulative: 123.1
    - (3) May/June 2013 (7.5 bn) — 17 May 2013: Disbursement (bn euros): 4.2 — Cumulative: 127.3; 25 June 2013: Disbursement (bn euros): 3.3 — Cumulative: 130.6
    - (4) July/December 2013 (3 bn) — 31 July 2013: Disbursement (bn euros): 2.5 — Cumulative: 133.1; 18 December 2013: Disbursement (bn euros): 0.5 — Cumulative: 133.6
    - (5) April/August 2014 (8.3 bn) — 28 April 2014: Disbursement (bn euros): 6.3 — Cumulative: 139.9; 9 July 2014: Disbursement (bn euros): 1.0 — Cumulative: 140.9; 14 August 2014: Disbursement (bn euros): 1.0 — Cumulative: 141.9
- 3rd programme (ESM):
  - Lasted August 2015 to August 2018.
  - Disbursed 61.9 bn ewith loan maturities ranging from 2034 to 2060 (32.5 years average loan maturity).
  - Table 20 (ESM disbursements by sub-tranche):
    - (1) August/December 2015 (18.7 bn) — 20 August 2015: Disbursement: 13.00 — Cumulative: 13.00; 24 November 2015: Disbursement: 2.00 — Cumulative: 15.00; 01 December 2015: Disbursement: 2.70 — Cumulative: 17.70; 08 December 2015: Disbursement: 2.70 — Cumulative: 20.40; 23 December 2015: Disbursement: 1.00 — Cumulative: 21.40
    - (2) June/October 2016 (10.3 bn) — 21 June 2016: Disbursement: 7.50 — Cumulative: 28.90; 26 October 2016: Disbursement: 2.80 — Cumulative: 31.70
    - (3) July/October 2017 (8.5 bn) — 10 July 2017: Disbursement: 7.70 — Cumulative: 39.40; 30 October 2017: Disbursement: 0.80 — Cumulative: 40.20
    - (4) March/June 2018 (6.7 bn) — 28 March 2018: Disbursement: 5.70 — Cumulative: 45.90; 15 June 2018: Disbursement: 1.00 — Cumulative: 46.90
    - (5) March/June 2018 (6.7 bn) — 06 August 2018: Disbursement: 15.00 — Cumulative: 61.90

*Contagion as a Dealmaker? The effect of financial spillovers on regional lending programs — Working Paper No. WP/22/133*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022133-print-pdf.pdf_