## pp073117-background-paper-collaboration-between-regional-financing-arrangements-and-the-imf

## Source details

**Canonical URL:** [pp073117-background-paper-collaboration-between-regional-financing-arrangements-and-the-imf](https://www.imf.org/-/media/files/publications/pp/2017/pp073117-background-paper-collaboration-between-regional-financing-arrangements-and-the-imf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/pp/2017/pp073117-background-paper-collaboration-between-regional-financing-arrangements-and-the-imf.pdf.md)
- [Structured JSON version](/-/media/files/publications/pp/2017/pp073117-background-paper-collaboration-between-regional-financing-arrangements-and-the-imf.pdf.json)

---

### Overview
- Purpose:
  - Provides background for Collaboration between Regional Financing Arrangements and the IMF (SM/17/190) by: (i) summarizing key characteristics of major regional financing arrangements (RFAs); (ii) presenting detailed case studies of selected past IMF-RFA co-financing episodes; and (iii) presenting technical details of the stylized contagion model used in the main paper.
- Chapter focus:
  - Institutional characteristics of RFAs (membership, resources, decision making, governance).
  - Lending features (instruments, objectives, modalities).
  - Operational capacities (surveillance, monitoring, information sharing).
  - Nature of conditionality and links with Fund support.
  - Case studies and technical contagion model details.

### RFAs covered (list)
- Arab Monetary Fund (AMF)
- BRICS’ Contingent Reserve Arrangement (CRA)
- Chiang-Mai Initiative Multilateralization (CMIM)
- Eurasian Fund for Stabilization (EFSD)
- European Union’s Balance of Payments facility (EU-BoP)
- European Financial Stability Mechanism (EFSM)
- European Stability Mechanism (ESM)
- Latin American Reserve Fund (FLAR)

### Key institutional and operational characteristics (cross-cutting observations)
- IMF linkage in facilities:
  - Many RFAs structure larger access or a majority of resources as IMF-linked (IMF-linked portion vs. IMF-delinked portion), commonly with a 30 percent quick-disbursing (IMF-delinked) share and the remainder linked to an IMF program (explicit for BRICS CRA, CMIM, and others).
- Information sharing:
  - Several RFAs report no formal mechanism for information sharing with the IMF in their treaties or regulations (AMF, BRICS CRA, CMIM, EFSD, FLAR, some EU arrangements).
  - Ad hoc MoUs and cooperative practices exist in some cases.
- Usage patterns:
  - Some RFAs never used their facilities (BRICS CRA, CMIM swap facilities).
  - Others have active lending histories (AMF, EFSD, FLAR, EFSF/ESM, EFSM, EU BoP facility).

### A. Arab Monetary Fund (AMF) — highlights
- Establishment: 1976.
- Membership: 22 members (listed in source).
- Resources:
  - AMF has SDR3.6 billion (US$4.8 billion) in capital as of end-2016, of which paid-in capital is SDR 2.5 billion (US$3.3 billion).
- Key lending facilities and access limits (examples):
  - Automatic Loan: access up to 75 percent of the member’s paid-in capital; maturity three years; not conditional on program implementation (subject to outstanding conditional loans rules).
  - Ordinary Loan: generally up to 100 percent of paid-in capital (can be supplemented to total 175 percent); requires macroeconomic adjustment program of at least one year.
  - Extended Loan: access up to 175 percent of paid-in capital (can be supplemented to total 250 percent); requires structural adjustment program of at least two years.
  - Oil Facility: access up to 200 percent of paid-in capital; up to 100 percent under simple and quick procedure without reform program.
  - Short-Term Liquidity Facility: access up to 100 percent of paid-in capital; extended promptly without prior agreement on a reform program.
- Fund role and information sharing:
  - No explicit institutional role for the IMF; several AMF facilities require members to first withdraw IMF reserve tranche and from similar organizations.
  - Information sharing: None (but MoU in April 2015 for training and joint analytical work).
- Recent activity (examples):
  - 2016 loans—Mauritania US$50 million (compensatory), Egypt US$82 million (compensatory), Jordan US$40 million (automatic).
  - Since establishment AMF extended 177 loans to 14 member states for a total of US$8.4 billion.

### B. BRICS Contingent Reserve Arrangement (CRA) — highlights
- Establishment timeline: dialogue origins 2006; treaty signed 2014; South Africa joined April 2011.
- Members: Brazil, China, India, Russia, South Africa.
- Resources: Size US$100 billion (about SDR74.4 billion). Commitments: China US$41 billion; South Africa US$5 billion; Brazil, Russia and India US$18 billion each.
- Key features:
  - Objectives: currency swaps to support each other for actual or potential short-term BoP pressures.
  - Access multiples by member commitment: China 0.5; Brazil, Russia, India 1; South Africa 2.
  - Fund role: 30 percent of maximum drawable amount available subject only to agreement of providing parties; remaining 70 percent linked to on-track IMF arrangement.
  - Surveillance: No surveillance and monitoring functions in CRA.
  - Information sharing: No formal mechanism in treaty.
- Recent activities: Never used.

### C. Chiang Mai Initiative Multilateralization (CMIM) — highlights
- Establishment: CMI launched May 2000; CMIM multilateral swap agreement effective March 24, 2010.
- Members: ASEAN10 + China, Japan, Korea; Hong Kong MA involvement.
- Resources: Size US$240 billion (about SDR178.5 billion); ten ASEAN members collectively US$48 billion; China, Japan, Korea and Hong Kong SAR US$192 billion.
- Facilities and access:
  - CMIM-PL and CMIM-SF; access determined by purchasing multiple (e.g., China 0.5; Vietnam 5.0 in examples).
  - Maturities: IMF-delinked portion 6 months; IMF-linked portion one year; supporting periods up to two or three years depending on facility.
  - Delinked portion increased from 20 to 30 percent in July, 2014.
- Surveillance and monitoring:
  - ASEAN+3 Macroeconomic Research Office (AMRO) established 2011; in 2016 became an international organization under the AMRO Agreement with monitoring and assessment mandates.
- Fund role:
  - 30 percent of maximum drawable amount categorized as quick disbursing (IMF de-linked portion); remaining portion linked to IMF program.
- Recent activities: Never used; regular test runs; first joint test run with IMF in 2016.

### D. Eurasian Fund for Stabilization and Development (EFSD) — highlights
- Establishment: June 9, 2009 (as EURASEC Anti-Crisis Fund).
- Members: Armenia, Belarus, Kazakhstan, Kyrgyz Republic, Russian Federation, Tajikistan.
- Resources: Budget contributions total US$8.5 billion (about SDR 6.3 billion).
- Lending instruments (examples):
  - Financial credits (FC): Stabilization Credits to low-income countries; Sovereign Loans to medium-income countries.
  - Investment loans (IL): min loan sizes US$30 million or US$10 million depending on GNI per capita.
  - Grants: US$0.5 million to US$5 million.
- Fund role and coordination:
  - FCs and ILs coordinated with IFIs and can be co-financed.
  - No formal treaty information-sharing mechanism with the Fund.
- Recent activities (examples):
  - FCs completed: Tajikistan 2010 US$70 million; Belarus 2011 US$3 billion.
  - Ongoing/ongoing at time of source: Armenia 2015 US$300 million; Belarus 2016 US$2 billion.

### E. EU regional RFAs — highlights
- EU Balance-of-Payments (BoP) Assistance Facility:
  - Establishment: Council Regulation No. 332/2002 of 18 February 2002.
  - Resources: Total outstanding amount limited to € 50 billion (about SDR 40 billion).
  - Typical facility amount limited to €12 billion in principal.
  - Fund role: No formal link but financing typically in close cooperation with IMF.
  - Recent activities: Loans to Romania, Latvia and Hungary.
- European Financial Stabilization Mechanism (EFSM):
  - Establishment: May 2010; resources via EC borrowing up to €60 billion (about SDR 48 billion).
  - Fund role: EFSM Regulation allows financing outside EU subject to IMF conditionality; activation expected with joint EU/IMF support.
  - Recent activities: EFSM loans extended to Ireland and Portugal co-financed with Fund (Fund provided about 1/3).
- EFSF and ESM:
  - ESM capital stock about €700 billion, of which €80 billion paid-in and €620 billion callable; effective lending capacity €500 billion with 75 percent still available.
  - Instruments include loans, primary/secondary market purchases, precautionary credit lines (PCCL, ECCL), loans for indirect bank recapitalization, and direct recapitalization (cap €60 billion).
  - Fund role: Fund participation sought on technical and financial levels; member requesting ESM assistance expected to apply to Fund where possible.
  - Recent activities: EFSF/ESM loans and indirect bank recapitalization facilities used. Four co-financed EU/IMF programs noted (Ireland, Portugal, Greece (second program), Cyprus under EFSF/ESM).

### F. Latin American Reserve Fund (FLAR) — highlights
- Establishment: Andean Reserve Fund 1978; transformed into FLAR in 1991.
- Members: Bolivia, Colombia, Ecuador, Costa Rica, Peru, Uruguay, Venezuela.
- Resources: Paid-in capital US$2.9 billion; 2016 maximum disbursement capacity over US$4.7 billion (about SDR 3.5 billion).
- Lending instruments and access limits (examples):
  - Balance of Payments credit: three-year maturity, one-year grace; access limit 2.5 times paid-in capital; maximum limit US$1.6 billion for larger members and US$820 million for smaller members.
  - Liquidity Credit: one-year maturity; access limit one time paid-in capital; maximum US$656 million (larger), US$328 million (smaller).
  - Contingency Credit: six-month maturity; access limit 2 times paid-in capital; maximum US$1.3 billion (larger), US$656 million (smaller).
- Surveillance: FLAR has surveillance function and economic secretariat.
- Fund role: No explicit IMF role in institutional setup; information sharing: None.
- Recent activities (examples):
  - Central Bank of Venezuela received US$482.5 million loan (3-year maturity) in July 2016.
  - FLAR extended US$617.58 million to Ecuador in 2014, following US$515 million in 2012.

### Case studies of IMF–RFA co-financing (selected)
- Countries analyzed: Armenia (2014 EFF), Cyprus (2013 EFF), Hungary (2008 SBA), Ireland (2010 EFF), Latvia (2008 SBA), Romania (2009, 2011, 2013 SBAs).
- Common lessons across case studies:
  - Effective division of labor reflecting comparative advantage can facilitate co-lending.
  - Timely requests for financing and prompt responses reduce costs and limit contagion.
  - Parsimonious and focused conditionality improves implementation capacity and program coherence.
  - Early engagement and consistent, coordinated public communication reduce market uncertainty and program risks.
  - Uncoordinated decisions or delays among partners can be costly (e.g., Cyprus delays; Latvia review timing tensions).

### Stylized contagion model — purpose and structure
- Purpose:
  - Quantifies contagion arising from large balance of payments (BoP) shortfalls from a globally or regionally systemic economy as the shock spreads to connected economies.
  - Compares differential effectiveness of alternative layers of the Global Financial Safety Net (GFSN).
- Network setup:
  - Countries connected via bilateral asset and trade positions measured in U.S. dollars.
  - Gross international reserves R evolve with current account surpluses, inflows from foreign assets, and payments on foreign liabilities.
  - Key equation structure (as provided):
    - ∆푅푖,푡 = ∑ 푎푖푗 푟푗,푡 ∆푒푗,푡 ∈ 푁퐷푡 푗≠푖 ⏟ Return on net foreign assets − ∑ 푎푗𝑖 푟𝑖,푡 ∆푒𝑖,푡 ⏟ Return on net foreign liabilities + ∑푇퐵푖푗 (∆푒푗,푡−1) 푗≠푖 ⏟ Trade inflows
  - Liquidity and default rules:
    - If 푅푖>0 then country honors external liabilities (non-defaulted set NDt).
    - If 푅푖=0 then country cannot make foreign payments → defaulted set Dt.
  - Contagion trigger:
    - Initial suspension of external payments by systemic country h: initial reserve loss for country i is ∆푹푰,푡 = −푎푖푕,푡 푟ℎ,푡 푒ℎ,푡.
  - Cascading default condition (as provided): ∀ 푖 ∈푁퐹푡 if ∑푟 aℎ𝑖 ℎ∈퐹𝑖 > 푅𝑖 → Country i defaults: i ∈ F_{t+1}.
- Role of GFSN and timing:
  - External financing via GFSN: 퐸퐹푖𝑡 = 퐸퐹푅퐹퐴;푖,푡 + 퐸퐹퐼푀퐹;푖,푡 subject to maximum access limits 퐸퐹̅̅̅̅퐼푀퐹;푖, 퐸퐹̅̅̅̅푅퐹퐴;푖.
  - Stigma/delay: model allows 퐸퐹퐼푀퐹;푖,푡 = 0 during delays to reflect stigma; RFAs often provide faster disbursements but cover fewer countries.

### Empirical implementation and calibration
- Dataset:
  - 62 countries representing 78 percent of global GDP.
  - Bilateral asset positions and trade flows matched to actual exposures as of 2016.
  - Asset data: interbank liabilities (BIS locational) and portfolio investment (CPIS).
  - Trade flows: Direction of Trade Statistics (DOTS).
- Calibration:
  - Quarterly frequency starting equivalent to the first quarter of 2017.
  - RFAs considered in simulations: CMIM, BRICS, FLAR, ESM.

### Simulation results — key findings (from illustrative exercise)
- Base outcomes:
  - If countries rely exclusively on their own reserves, roughly 20 countries would be affected by contagion following the shocks.
  - Immediate access to Fund resources reduces affected countries to 9 in the case of a globally systemic shock.
  - Collaboration between the IMF and RFAs can halve the number of contagion events compared with Fund-only or RFA-only responses.
- Table 6 — The Effectiveness of the GFSN (Number of countries affected by the initial shock)
  - Columns (in order): IMF with no lag | IMF with 2-quarter lag | RFA only | IMF (no lag) + RFA | IMF (2-quarter lag) + RFA
  - Global shock: 9 | 19 | 13 | 5 | 12
  - Regional shock: 0 | 5 | 7 | 0 | 2
- Role of RFAs:
  - RFAs alone can materially reduce contagion for regional shocks, especially if the largest RFA members are not hit or the shock is not synchronized across RFA members.
  - RFAs can reduce contagion to 13 countries for a global initial shock, and reduce contagion by more than half for a regional shock.
- Importance of speed of IMF intervention:
  - The Fund’s effectiveness depends critically on timing; a 2-quarter delay (stigma) substantially weakens IMF-only outcomes and reduces the marginal benefit of IMF+RFA collaboration.

### Robustness and sensitivity
- Robustness:
  - Conclusions are qualitatively robust across different parameter values; relative performance ranking of GFSN constellations remains unchanged in robustness checks.
- Sensitivity to initial shock size:
  - Number of defaults depends crucially on assumed shock size (initial shock set equal to implied quarterly interest payments due).
  - Larger shocks can increase defaults more than proportionally; credit line size relative to shock matters.
  - Non-linear amplification: a single additional country falling under contagion can rapidly increase stress via multiple interactions, though relative performance ranking of GFSN constellations remains robust.

### Selected operational lessons from case studies (policy-relevant)
- Timeliness:
  - Prompt requests for financing and rapid provision of support are critical to limit contagion and program costs (illustrated by Hungary 2008 and Cyprus 2013 contrasts).
- Conditionality design:
  - Parsimonious and focused conditionality enhances implementation capacity and program coherence (repeated across Cyprus, Ireland, Romania, Latvia).
- Division of labor:
  - Effective cooperation reflects comparative advantage (Fund global resources and technical capacity; RFAs speed, regional knowledge, and sometimes financing windows).
- Coordination and communication:
  - Early engagement and coordinated public communication reduce stigma and market uncertainty; uncoordinated actions or review timing can undermine program credibility (Latvia example).
- Information sharing and formal frameworks:
  - Lack of formal information-sharing mechanisms across many RFAs can complicate prompt collaborative responses; ad hoc MoUs have been used in some cases (AMF MoU April 2015).

*Source: pp073117-background-paper-collaboration-between-regional-financing-arrangements-and-the-imf — Chapter 4 (as provided).*

### 1. AMF and the IMF Programs (2000–16) _____________________________________________ 7

### 1. AMF and the IMF Programs (2000–16)

### Overview
- Purpose of the chapter:
  - Provides background information for Collaboration between Regional Financing Arrangements and the IMF (SM/17/190) by: (i) summarizing the key characteristics of major regional financing arrangements (RFAs); (ii) presenting detailed case studies of selected past IMF-RFA co-financing episodes; and (iii) presenting the technical details of the stylized contagion model used in the main paper.
- Structure highlights:
  - Summarizes institutional characteristics of RFAs covered.
  - Discusses lending instruments, objectives, lending modalities, surveillance and monitoring capacity, information-sharing procedures, and the nature of conditionality and links with Fund support.
  - Presents case studies of recent IMF-RFA co-financing episodes and draws lessons for operational principles.
  - Provides technical details of the contagion model used in the main paper.

### RFAs covered
- Arab Monetary Fund (AMF)
- BRICS’ Contingent Reserve Arrangement (CRA)
- Chiang-Mai Initiative Multilateralization (CMIM)
- Eurasian Fund for Stabilization (EFSD)
- European Union’s Balance of Payments facility (EU-BoP)
- European Financial Stability Mechanism (EFSM)
- European Stability Mechanism (ESM)
- Latin American Reserve Fund (FLAR)

### Chapter content and focus areas
- Institutional characteristics described:
  - Membership
  - Resources
  - Decision making process
  - Governance
- Lending features described:
  - Lending instruments (facilities)
  - Objectives
  - Lending modalities
- Operational capacities discussed:
  - Surveillance and monitoring capacity
  - Procedures for information sharing
- Conditionality:
  - Examination of the nature of conditionality each RFA requires
  - Whether the RFA has any link with Fund support
- Recent activity:
  - Brief overview of recent activity for each RFA
- Supporting materials:
  - Information builds on available public information and supports Table 1 and Figure 5 in the main paper.

### Case studies of past IMF–RFA co-financing (listed)
- Armenia (2014 EFF)
- Cyprus (2013 EFF)
- Hungary (2008 SBA)
- Ireland (2010 EFF)
- Latvia (2008 SBA)
- Romania (2009, 2011, 2013 SBAs)

- Sources for case studies:
  - Official Fund and RFA reports
  - Inputs from Fund staff
  - Bilateral discussions with individual RFAs

### Tables and annexes referenced in the chapter
- Tables (selection from chapter front matter):
  - CMIM: Financial Resources and Access (Table 1)
  - Programs under the EU’s Balance of Payment Facility (Table 2)
  - Programs under the EFSM (Table 3)
  - Programs under the EFSF and ESM (Table 4)
  - Country Sample for Network Model (Table 5)
  - The Effectiveness of the GFSN (Table 6)
- Other listings in the chapter front matter:
  - Loans by FLAR and the IMF since 1978
  - Cyprus—Total Customer Deposits
  - Cyprus—Structural Conditionality
  - Latvia––Macroeconomic Adjustment
  - References

*International Monetary Fund — COLLABORATION WITH RFA AND THE IMF—BACKGROUND PAPER*

### 4. Cross-Country Contagion and the Global Financial Safety Net (GFSN). The stylized

### 4. Cross-Country Contagion and the Global Financial Safety Net (GFSN)

### Stylized contagion model
- Purpose: Quantifies contagion arising from large balance of payments (BoP) shortfalls emanating from a globally or regionally systemic economy as the shock spreads to other connected economies.
- Use: Analytical framework allows comparison of the differential effectiveness of alternative layers of the GFSN in stemming contagion.
- Application: Model described in this chapter is used to generate the results reported in Box 1 of the main paper.

### Main features of major Regional Financing Arrangements (RFAs)

- Overview: The chapter provides standardized descriptions of establishment, membership, resources, governance and decision making, lending instruments (objectives, facilities and modalities), surveillance and monitoring, the Fund’s role in lending operations, conditionality, information sharing procedures with the IMF, and recent activities.

### A. Arab Monetary Fund (AMF)
- Establishment: 1976.
- Members: Algeria, Bahrain, Comoros, Djibouti, Egypt, Iraq, Jordan, Kuwait, Lebanon, Libya, Mauritania, Morocco, Oman, Palestine, Qatar, Saudi Arabia, Somalia, Sudan, Syria, Tunisia, United Arab Emirates, and Yemen.
- Resources: AMF has SDR3.6 billion (US$4.8 billion) in capital as of end-2016, of which paid-in capital is SDR 2.5 billion (US$3.3 billion). The AMF may borrow and can issue securities if decided by the Board of Governors.
- Governance and decision making: Director General, Chairman of the Board of Executive Directors, six specialized executive management departments, and executive management committees (Loan, Investment, Administration and Risk Management Committees).
- Lending instruments (objectives and facilities/modalities):
  - Objectives: Correct BoP disequilibria, remove restrictions on current payments among members, advise on investment of member financial resources in foreign markets, promote Arab financial markets, and foster policies toward a unified Arab currency.
  - Automatic Loan: access up to 75 percent of the member’s paid-in capital in convertible currencies; maturity of three years; not conditional on program implementation (subject to outstanding conditional loans rules).
  - Ordinary Loan: generally up to 100 percent of paid-in capital, could be supplemented with an Automatic Loan (total 175 percent); must be accompanied by a macroeconomic adjustment program of at least one year.
  - Extended Loan: access up to 175 percent of paid-in capital, can be supplemented by an Automatic Loan to access a total of 250 percent of paid-in capital; requires structural adjustment program of at least two years.
  - Compensatory Loan: access up to 100 percent of paid-in capital; designed for unanticipated BoP needs from export shortfalls or increased agricultural imports.
  - Structural Adjustment Facility: access up to 175 percent of paid-in capital for each sector; for members with progress in macroeconomic stability and sectoral reform programs.
  - Trade Reform Facility: up to 175 percent of paid-in capital to assist trade reform implementation costs.
  - Oil Facility: access up to 200 percent of paid-in capital; up to 100 percent under simple and quick procedure without reform program; full amount requires accompanying reform supported by ordinary facilities.
  - Short-Term Liquidity Facility: access up to 100 percent of paid-in capital; extended promptly without prior agreement on a reform program.
- Surveillance and monitoring: Consultations to monitor adjustment programs during loan maturity; technical assistance and capacity building through the Economic Policy Institute.
- Fund’s role: No explicit institutional role for the IMF, but several AMF facilities require members to first withdraw IMF reserve tranche and from similar organizations.
- Conditionality: Varies by facility (see facilities above for specifics).
- Information sharing (with the Fund) procedure: None. (Footnote: AMF and the Fund signed an MoU in April 2015 for training, capital markets development, Arabstat, and joint analytical work.)
- Recent activities: 2016 loans—compensatory loans to Mauritania (US$50 million) and Egypt (US$82 million), and an automatic loan to Jordan (US$40 million). Egypt received an Automatic Loan of US$250 million in 2015; Sudan received an Extended Loan in 2015–16 of US$166 million; Jordan received Structural Adjustment Facility US$55.4 million (June 2015–June 2016); Egypt received Structural Adjustment Facility US$330 million (May 2015–May 2016). Since establishment AMF extended a total of 177 loans to 14 member states for a total of US$8.4 billion.

### B. BRICS Contingent Reserve Arrangement (CRA)
- Establishment: Dialogue origins in 2006; First Summit 2009; South Africa joined April 2011; treaty establishing CRA signed in 2014.
- Members: Brazil, China, India, Russia, and South Africa.
- Resources: Size US$100 billion (about SDR74.4 billion). China US$41 billion; South Africa US$5 billion; Brazil, Russia and India US$18 billion each.
- Governance and decision making: Governing Council and Standing Committee; Governing Council works by consensus on strategic issues; Standing Committee for executive/operational decisions; chairing party acts as coordinator; lending decisions by simple majority weighted by financial commitments; waivers and sanctions by consensus of providers. CRA currently does not possess independent international legal personality.
- Lending instruments (objectives and facilities/modalities):
  - Objectives: Commitments to support each other through currency swaps in case of actual or potential short-term BoP pressures; ability to swap domestic currencies for USD.
  - Facilities: liquidity instrument (short-term BoP pressures) and precautionary instrument (potential short-term BoP pressures).
  - Access: determined by multiple of member’s commitment: China 0.5; Brazil, Russia, India 1; South Africa 2.
  - Maturities: 6 months with up to 3 renewals for IMF-delinked portion; one year with up to 2 renewals for IMF-linked portion.
  - Interest: benchmark rate for corresponding maturity plus a spread that increases periodically up to a limit; precautionary instrument commitment fee on undrawn amounts.
- Surveillance and monitoring: No surveillance and monitoring functions.
- Fund’s role in lending operations: 30 percent of maximum drawable amount available subject only to agreement of providing parties; remaining 70 percent linked to an on-track IMF arrangement with financing based on conditionality.
- Conditionality: Limited document/data submission requirements; compliance with Treaty; no arrears with parties or public financial institutions; no arrears with multilateral and regional financial institutions; compliance with surveillance and information obligations to the IMF.
- Information sharing (with the Fund) procedure: No formal mechanism specified in treaty.
- Recent activities: Never used.

### C. Chiang Mai Initiative Multilateralization (CMIM)
- Establishment: Launched May 2000 as CMI; CMIM multilateral swap agreement took effect March 24, 2010; CMIM replaced network of BSAs but does not preclude separate bilateral swaps; ASA remained in effect.
- Members: ASEAN countries (Brunei Darussalam, Cambodia, Indonesia, Lao PDR, Malaysia, Myanmar, Philippines, Singapore, Thailand, and Vietnam) and Plus Three countries (China, Japan, and Korea) and Hong Kong Monetary Authority involvement.
- Resources: Size US$240 billion (about SDR178.5 billion). Ten ASEAN members collectively contribute US$48 billion; China, Japan, Korea and Hong Kong SAR collectively contribute US$192 billion.
- Governance and decision making: Swap activation initiated by any CMIM Party to CMIM Coordinating Countries (two Chairs). Coordinating Countries deliver request to Executive Level Decision Making Body (ELDMB) which decides on swap request. Decision process to be completed within two weeks of swap request notice delivery to ELDMB.
- Lending instruments (objectives and facilities/modalities):
  - Objectives: Provide USD financial support through currency swap transactions to address potential and actual BoP and short-term liquidity difficulties and supplement international arrangements.
  - Facilities: CMIM Precautionary Line (CMIM-PL) and CMIM Stability Facility (CMIM-SF).
  - Access: Determined by purchasing multiple of each member’s commitment (examples in Table 1; e.g., China 0.5 purchasing multiple; Vietnam 5.0).
  - CMIM-PL maturities: 6 months (IMF-delinked portion) and one year (IMF-linked portion); maximum supporting period two years.
  - CMIM-SF maturities: 6 months (IMF-delinked) and one year (IMF-linked); maximum supporting period 2 years (IMF-delinked) and 3 years (IMF-linked).
- Surveillance: ASEAN+3 Macroeconomic Research Office (AMRO) established 2011 as independent regional surveillance unit; AMRO conducts annual consultations, prepares quarterly consolidated reports, and in 2016 became an international organization under the AMRO Agreement with defined mandates including monitoring, assessment, identification of risks, and support for implementation of CMIM.
- Fund’s role in lending operations: 30 percent of maximum drawable amount categorized as quick disbursing (IMF de-linked portion); remaining portion linked to IMF program.
- Monitoring function: AMRO prepares recommendations on swap requests, monitors use and impact of funds, and provides borrower review to creditor satisfaction. Requirements can be waived by 2/3 of effective votes of ELDMB.
- Information sharing (with the Fund) procedure: No formal mechanism in place.
- Recent activities: Never used; members have conducted regular test runs. First joint test run with the IMF conducted in 2016. (Note: delinked portion increased from 20 to 30 percent in July, 2014.)

### D. Eurasian Fund for Stabilization and Development (EFSD)
- Establishment: First established June 9, 2009 as the EURASEC Anti-Crisis Fund.
- Members: Armenia, Belarus, Kazakhstan, Kyrgyz Republic, Russian Federation, and Tajikistan.
- Resources: Budget contributions total US$8.5 billion (about SDR 6.3 billion).
- Governance and decision making: EFSD Council (Ministers of Finance) makes key decisions; chaired by the Russian Federation Finance Minister.
- Lending instruments (facilities/modalities):
  - Financial credits (FC): support anti-crisis and stabilization programs; two loan types:
    - Stabilization Credits to Low-Income Countries: up to 20 years’ maturity, up to 5 years’ grace; fixed interest rate 1–3 percent; program duration up to 3 years.
    - Sovereign Loans to Medium-Income Countries: up to 10 years’ maturity, up to 5 years’ grace; interest rate six months-LIBOR+Russia/Kazakh spread.
    - Size within country access limits; extended in USD and Euro only; minimum FC size US$10 million.
  - Investment loans (IL): for member states or companies implementing interstate investment projects; minimum loan size US$30 million for countries with GNI per capita over US$5,000 and US$10 million for countries with GNI per capita under US$5,000; interest six-month LIBOR+Russia/Kazakh spread; up to 15 (10) years maturity and 5-year grace for governments (companies). ILs to low-income countries carry grant element no less than 35 percent.
  - Grants: finance social sector programs; grant size US$0.5 million to US$5 million with 1.5 years implementation.
- Surveillance and monitoring: Policy/project implementation measured by specific indicators and evaluated by EFSD Council based on reports from EFSD Manager.
- Fund’s role: FCs coordinated with IFIs, bilateral donors and can be co-financed; ILs coordinated and can be co-financed.
- Conditionality: FCs require an Anti-Crisis Program by the government; ILs provided only if no other financing source available on acceptable terms; security requirements for ILs include sovereign guarantees, rated financial institution guarantees, highly liquid assets acceptable to EFSD Council, or third party suretyships.
- Information sharing (with the Fund) procedure: No formal mechanism in treaty.
- Recent activities:
  - Financial Credits: Completed—Tajikistan 2010 US$70 million; Belarus 2011 US$3 billion. Ongoing—Armenia 2015 US$300 million; Belarus 2016 US$2 billion.
  - Investment loans: Projects including Bishkek-Osh Road Rehabilitation Phase IV (Kyrgyz Republic), agricultural machinery supplies (Kyrgyz Republic), North-South Road Corridor (Armenia), Toktogul HPP Rehabilitation (Kyrgyz Republic), Irrigation System Modernization (Armenia).

### E. RFAs in the European Union
- Several EU arrangements described: Balance-of-Payments (BoP) Assistance Facility, European Financial Stabilization Mechanism (EFSM), European Financial Stability Facility (EFSF) and European Stability Mechanism (ESM).

  - Balance-of-Payments (BoP) Assistance Facility
    - Establishment: Council Regulation No. 332/2002 of 18 February 2002; replaces earlier 1988 facility.
    - Members: Non-euro EU member states: Bulgaria, Croatia, Czech Republic, Denmark, Hungary, Poland, Romania, Sweden, and the United Kingdom.
    - Resources: Total outstanding amount limited to € 50 billion (about SDR 40 billion) in principal.
    - Governance and decision making: Request presented to the European Commission (EC); EU Council decides based on EC recommendation.
    - Lending instruments/objectives: Mutual assistance to non-euro area Member States in BoP difficulties; assistance designed to ease external financing constraints and limited to €12 billion in principal.
    - Facilities/modalities: EU Council decides loan/financing facility amount and average duration (normally about five years); ‘AAA’ loan rates passed on without additional margin.
    - Surveillance and monitoring: EU institutions assess compliance; post-program surveillance resembling Fund’s PPM.
    - Fund’s role: No formal link to a Fund-supported program, but recent assistance extended in close cooperation with IMF and others.
    - Conditionality: Member presents a draft adjustment program; typical conditions include fiscal consolidation, governance, financial sector stabilization, structural reforms, and safeguards against fraud; precautionary arrangements include ex-post conditionality.
    - Information sharing (with the Fund): No formal mechanism; EU-BoP financing typically occurs in cooperation with the Fund.
    - Recent activities: Loans to Romania, Latvia and Hungary (details in Table 2).

  - European Financial Stabilization Mechanism (EFSM)
    - Establishment: May 2010 by Council Regulation (EU) No 407/2010; amended August 2015 by Council Regulation (EU) 2015/1360.
    - Members: All EU member states.
    - Resources: EC authorized to borrow; maximum financing capacity €60 billion (about SDR 48 billion) backed by implicit EU guarantee.
    - Governance and decision making: Council decision by qualified majority on EC proposal; EC (in consultation with ECB) reexamines conditionality during program.
    - Lending instruments/objectives: External financial assistance for member states experiencing or threatened by severe economic/financial disturbance beyond control.
    - Facilities/modalities: Amount, duration and disbursement terms decided by Council; MoU and Loan Agreement set terms and program conditions; maturities in past ranged 7 to 30 years; bridge loan to Greece July 2015 had maturity 1 month.
    - Surveillance and monitoring: EC (in consultation with ECB) designs and monitors adjustment program at least every six months.
    - Fund’s role: EFSM Regulation allows financing outside EU subject to IMF conditionality; activation expected in context of joint EU/IMF support.
    - Conditionality: Beneficiary implements policy conditions to re-establish sound economic/financial situation and restore market financing capacity.
    - Information sharing (with the Fund): No formal mechanism in regulation.
    - Recent activities: EFSM loans extended to Ireland and Portugal co-financed with Fund (Fund provided about 1/3). Table 3 shows Ireland 85.0/22.5 (Jan 2011–Nov 2013), Portugal 78.0/24.0 (May 2011–May 2014), Greece 7.1/7.1 (July–Aug 2015 bridge loan, fully repaid).

  - European Financial Stability Facility (EFSF) and European Stability Mechanism (ESM)
    - Establishment: EFSF created June 2010 as temporary crisis mechanism; since 2012 ESM performs financial assistance role; EFSF continues to manage prior loans and bond operations.
    - Members: Euro area member states.
    - Resources: ESM capital stock about €700 billion, of which €80 billion paid-in and €620 billion callable; effective lending capacity €500 billion with 75 percent still available.
    - Governance and decision making: ESM intergovernmental organization; Board of Governors (Finance Ministers) generally require unanimity for key decisions; Board of Directors appointed by Governors; EC delegated authority for debt sustainability analysis, negotiate MoUs, and monitor programs.
    - Lending instruments:
      - Loans to member states within macroeconomic adjustment programs.
      - Primary market purchases of bonds or debt securities at market prices.
      - Secondary market purchases to support government debt market functioning.
      - Precautionary credit lines: Precautionary Conditioned Credit Line (PCCL) and Enhanced Conditions Credit Line (ECCL).
      - Loans for indirect bank recapitalization (loans to government for bank recapitalization).
      - Direct recapitalization of institutions (inject capital directly into financial institutions).
    - Modalities:
      - Precautionary credit lines initial availability one year, renewable twice for six months each.
      - Primary market purchases normally limited to maximum 50 percent of final issued amount.
      - Direct recapitalization instrument cap on total available resource €60 billion.
      - No access limits set for loans to individual members; access based on financing need and subject to MoU containing conditionality.
    - Surveillance and monitoring: Delegated to EC; program monitoring liaises with ECB and Fund; debt sustainability analysis jointly conducted by EC and Fund (Article 13/1).
    - Fund’s role: Fund participation sought on technical and financial levels; member requesting ESM assistance expected to, where possible, apply to Fund.
    - Conditionality: Varies by instrument; loans require macroeconomic adjustment program; primary/secondary purchases and credit lines tied to MoU and eligibility criteria; bank recapitalization instruments carry financial-sector-specific conditionality.
    - Information sharing (with the Fund): No specific ESM-IMF provisions, but beneficiary member state obliged to supply IMF-dispatched documents while any amount outstanding.
    - Recent activities: EFSF/ESM loan and indirect bank recapitalization facilities have been used. Precautionary credit lines, primary and secondary market purchases, and direct recapitalization instruments have not been used. Four co-financed EU/IMF programs to date: Ireland, Portugal, Greece (second program) and Cyprus under EFSF/ESM. Table 4 lists program totals and statuses (e.g., Greece 172.7/144.7 March 2012–December 2014 Expired June 2015; Greece 86.0/86.0 July 2015–August 2018 Active with €31.7bn disbursed from ESM).

### F. Latin American Reserve Fund (FLAR)
- Establishment: Andean Reserve Fund established 1978; transformed into FLAR in 1991.
- Members: Bolivia, Colombia, Ecuador, Costa Rica, Peru, Uruguay, and Venezuela.
- Resources: Paid-in capital US$2.9 billion. In 2016 maximum disbursement capacity over US$4.7 billion (about SDR 3.5 billion).
- Governance and decision making: Assembly of Representatives (Finance Ministers) is highest authority; rotating President one-year term; each member has a single vote. Board composed of Central Bank Governors and an Executive President; Executive Presidency is main permanent technical body.
- Lending instruments (objectives and facilities/modalities):
  - Objectives: Provide BoP support through loans/credit guarantees; foster harmonization of exchange rate, monetary and financial policies; improve international reserve investment conditions.
  - Balance of Payments credit: three-year maturity, one-year grace; access limit 2.5 times paid-in capital; maximum limit US$1.6 billion for larger members (Colombia, Peru, Venezuela) and US$820 million for smaller members (Bolivia, Costa Rica, Ecuador, Uruguay); interest 3-month Libor plus 300–400 bps; prepaid charge/commission 30 bps.
  - Liquidity Credit: one-year maturity; access limit one time paid-in capital; maximum lending limit US$656 million for larger members and US$328 million for smaller members; interest 3-month Libor plus 170–230 bps; prepaid charge/commission 10 bps.
  - External Debt restructuring of Central Banks: three-year maturity, one-year grace; access limit 1.5 times paid-in capital; maximum access US$984 million for larger members and US$492 million for smaller members; interest 3-month Libor plus 300–400 bps; prepaid charge/commission 30 bps.
  - Contingency Credit: six-month maturity; access limit 2 times paid-in capital; maximum limit US$1.3 billion for larger members and US$656 million for smaller members; interest 3-month Libor plus 140–190 bps; prepaid charge/commission 10 bps.
  - Treasury Operations: financing up to 30 days; access limit 2 times paid-in capital.
- Surveillance and monitoring: FLAR has surveillance function and an economic secretariat to monitor lending operations.
- Fund’s role in lending operations: No explicit IMF role envisaged in institutional set up.
- Conditionality: Central bank must provide report on monetary, credit, exchange, fiscal and trade policies to correct disequilibria; FLAR staff presents report to Board; historically FLAR has always supported member requests and members have repaid on time (sometimes early).
- Information sharing (with the Fund) procedure: None.
- Recent activities: Central Bank of Venezuela received US$482.5 million loan (3-year maturity) in July 2016. FLAR extended US$617.58 million to Ecuador in 2014, following US$515 million in 2012. FLAR provides technical assistance (e.g., 2015 to central banks of Venezuela and Peru; support to Colombia’s Finance Ministry).

### Cross-cutting observations from RFA descriptions
- IMF linkage in facilities: A common feature is that many RFAs structure larger access or a majority share of resources as IMF-linked (IMF-linked portion vs. IMF-delinked portion), often with a 30 percent quick-disbursing (IMF-delinked) share and the remainder linked to IMF program conditionality (explicit for BRICS CRA, CMIM, CMIM/AMRO, and others).
- Information sharing: Several RFAs report no formal mechanism for information sharing with the IMF in their treaties or regulations (AMF, BRICS CRA, CMIM, EFSD, FLAR, some EU arrangements), though ad hoc MoUs and cooperative practices exist in some cases.
- Usage: Some regional arrangements have never been used (BRICS CRA, CMIM swap facilities), while others have active lending histories (AMF, EFSD, FLAR, EFSF/ESM, EFSM, EU BoP facility).

*Source: pp073117-background-paper-collaboration-between-regional-financing-arrangements-and-the-imf — Chapter 4 (as provided).*

### 5. In July 2014, the IMF Board approved an arrangement under the Extended Fund

### 5. In July 2014, the IMF Board approved an arrangement under the Extended Fund

### A. EFSD involvement and objectives
- In July 2014, the IMF Board approved an arrangement under the Extended Fund Facility for SDR 82.21 million (89 percent of quota).
- Aim: to consolidate economic stability—following 2010–13 arrangements under the EFF and EFC—and rebuild buffers against possible external shocks.
- Arrangement also aimed at supporting further reforms to facilitate the transition towards a dynamic emerging market economy.
- In October 2015, the EFSD Council approved a US$300 million budget support loan in support of the goals set out in the authorities’ EFF program.
- Disbursements were designed to ensure support for specific activities under the Stimulus Program to promote economic growth and boost export-driven industries.
- EFSD involvement helped limit program risks through the provision of budget financing and more generally supporting the program’s objectives.
- The EFSD drew implications from the Fund’s program design and macroeconomic framework when setting its conditionality (EFSD, 2015a, b).
- At an operational level the EFSD-Fund collaboration was open, cordial, and effective, although there was no formal agreement on Fund-EFSD coordination.

Lessons for cooperation
- Effective cooperation and division of labor, reflecting each institution’s comparative advantage, can facilitate co-lending.

*This case study benefited from EFSD documents and discussions with Fund staff.*

### B. Cyprus (2013 EFF) — case overview
- Purpose: assess operational efficiency of cooperation between the Fund and European partners (EC/ECB/ESM) in the Cyprus EFF; illustrate costs of not requesting financing assistance timely and difficulties in reaching agreement among lenders.
- Once the program commenced, collaboration was effective throughout and the adopted program strategy was highly successful, with quicker recovery, faster adjustment, and an early exit from the program.

Key chronology and facts
- Imbalances culminating in intervention in major domestic banks were evident in late-2011, but authorities requested assistance from the EU and the IMF in mid-2012.
- Contributing factors: imbalances accumulated before the GFC, lax fiscal policy response to the GFC, deep financial links to Greece, explosion of Cyprus’s main power station, restructuring of Greek sovereign debt (IMF, 2014).
- Authorities secured a loan from Russia in late 2011, providing temporary support without conditionality.
- Continuous deposit outflows and tight sovereign-bank links resulted in collapse of major banks and sharp deterioration of public finances; authorities requested official assistance from the EU and the IMF in mid-2012.
- Almost a year elapsed between the request for financial assistance and the announcement of a staff-level agreement—contrasting with Greece (2010), Ireland (2010), and Portugal (2011) where agreement was reached within two to four weeks.
- The extended delay was in part due to a fundamental difference between the IMF and the EC/ECB/ESM related to the banking sector strategy:
  - The Fund supported resolution of insolvent banks, including bail-in of private creditors, aimed at restoring financial stability and limiting risk of an unsustainable debt path.
  - European partners opposed resolution of these banks given concerns about potential intra-euro area spillover risks and considered the IMF’s strategy highly risky.
- Impending presidential election in early 2013 and authorities’ opposition to some proposed conditionality also added to the delay.
- By March 2013, rapidly growing deposit outflows imposed additional costs; solution required larger bail-in of private creditors and substantially higher emergency liquidity assistance (ELA) to the banks.
- Inability to create a bad bank limited options for bank resolution and added program risks.
- Leaked information on banking strategy considerations hurt market confidence and accelerated deposit outflows.
- The EC’s structural conditions were far more numerous and detailed than those in the Fund program; this contributed to longer program discussions and strained authorities’ implementation capacity.
- Some proposed reforms (e.g., healthcare) had macro-fiscal costs affecting coherence and consistency of program design.
- Outcome: adjustment program delivered a strong turnaround—growth resumed faster than expected, fiscal position restored to a sustainable path, market access reestablished, banking system strengthened, NPL workouts accelerated, creating space for new lending.
- Authorities requested cancellation of the arrangement two months early.

Lessons for cooperation
- While delays can be costly, sustainable agreement over program issues must ensure a coherent program design that addresses the economic problems and safeguards, and has buy-in from the respective parties.
- Parsimonious and focused conditionality avoids straining the authorities’ implementation capacity and is important to ensure coherence and consistency of program design.
- Effective and consistent public communication reduces program uncertainty and limits the risks of financial market disruptions.

### C. Hungary (2008 SBA) — case overview
- First program co-financed by both the EU and the Fund; created a precedent for effective cooperation in program design and communication.
- Illustrated the Fund’s readiness to take the lead in designing an adjustment program at short notice.

Key chronology and facts
- In 2008, amid a rapidly spreading global financial crisis, Hungary requested a Stand-by Arrangement for SDR 10.5bn (about 12.5bn euros) with the Fund.
- As an EU member, Hungary was required to consult with the EU on BoP needs before requesting assistance from the Fund and other sources.
- Given crisis speed, the EU agreed to cooperate and provide co-financing with the Fund under accelerated procedures.
- Timely request by authorities and prompt response by Fund and EU with a large, front-loaded financing package were critical to stabilizing market confidence.
- Funding pressures receded, parent banks supported their subsidiaries, and government tapped international markets earlier than expected.
- The program was also the EU’s first operation under its BoP instrument.
- Financing package designed within four weeks and relied on exceptional and frontloaded access to Fund resources plus EU support.
- Objectives: prevent large capital outflows, restore confidence, reduce regional contagion risk; SBA referenced need to reduce financial spillovers.
- Full package reached 20bn euros, with a €6.5 billion contribution from the EU (about half of the Fund contribution).
- Program involved exceptional access for the Fund at 1,015 percent of quota.
- The EC had to adjust practice to the emergency nature of the operation; timing of disbursements and repayments was complementary (IMF frontloading, EC longer repayment window).
- Cooperation on the ground included attempts to align press releases, reviews, and disbursement cycles.
- Conditionality was set out in two distinct documents (MOU and LOI) with different phrasing; this did not complicate implementation or achievement of conditionality.
- Consistent communication timing was important given financial market sensitivity.

Lessons for cooperation
- The Fund demonstrated comparative advantage in responding to emergency situations and operational experience.
- Effective burden sharing is possible: Fund front loading and EC offering longer term support.
- Effective, consistent, and coordinated communication is important for program success.

### D. Ireland (2010 EFF) — case overview
- Aims: assess operational efficiency and modes of cooperation (and resolution of differences) between IMF and European partners (EC/ECB/ESM).
- Illustrates importance of authorities’ program ownership, parsimony in conditionality, and early engagement between program partners.

Key chronology and facts
- Staff-level agreement between IMF/EC/ECB and Irish authorities reached within a few weeks of the request for financial assistance.
- Irish authorities requested assistance on November 10, 2010; staff-level agreement announced on November 28, 2010.
- Efficiency aided by earlier preparations by Irish authorities, including a detailed National Recovery Program prepared in late 2010.
- Serious differences arose during program design regarding bail-in of senior bond holders:
  - Fund staff recommended burden sharing by senior bank creditors to reduce recapitalization expenses, fiscal adjustment needs, and support debt sustainability.
  - European authorities opposed bail-in due to potential cross-border spillovers to bank funding in the euro area.
  - Fund staff considered spillover risks manageable and partly mitigated by pricing and investor composition.
  - The Irish government supported bail-in but required agreement of European partners.
  - Final outcome: Fund agreed to proceed without burden sharing by senior bondholders but left open possibility of bailing-in these bonds in future reviews if needed.
- Inability to reach final agreement on burden sharing led partners to pursue alternative risk mitigation options:
  - Joint projects included developing an exit strategy from Emergency Liquidity Assistance provided by the Central Bank of Ireland.
  - Technical work led to an extension of maturity of EFSM loans by seven years in April 2013.
- Other differences (pace of fiscal consolidation, bank deleveraging) were resolved at technical level after extensive discussions:
  - European partners, especially the ECB, favored more front-loaded fiscal consolidation; they agreed to extend timeframe for reaching 3 percent deficit ceiling by one year.
  - ECB favored rapid bank deleveraging with proceeds used to repay ECB liquidity support; discussions led to phased sales focused on banks’ foreign assets with high transparency and safeguards against “fire sales.”
- EC’s conditionality was not as expansive as in other EA programs, reflecting limited need for structural reforms outside financial sector; EC conditionality focused on competitiveness and employment; IMF conditionality remained focused on financial sector reforms.
- Ireland’s Ex-post Evaluation noted that “a close and early Fund engagement with monetary union authorities can facilitate program success” (IMF, 2015, ¶65).
- There were few Ireland-specific interactions between Fund and European partners until late 2010, about two years after the property bubble burst; earlier close cooperation could have facilitated faster policy solutions.

Lessons for cooperation
- Strong ownership and capacity for program design and implementation by the authorities can anchor Fund-RFA cooperation.
- Parsimonious and focused conditionality increases probability of program success.
- Early engagement by the Fund with RFA authorities can facilitate program success in the event of a lending program.

### E. Latvia (2008 SBA) — case overview
- An early example of EU-IMF co-financing.
- Unique with respect to program exit strategy (euro adoption).
- Successful example where the Fund’s initial strategy was adjusted, reflecting ongoing discussions with the RFA and the authorities.
- Alternative strategy reflecting authorities’ policy preference (to maintain the peg) resulted in a consistent solution to Latvia’s problems.
- Illustrates how uncoordinated communications and actions can undermine parties and put the program at risk.

*COLLABORATION WITH RFA AND THE IMF––BACKGROUND PAPER*

### 22. The SBA for Latvia, approved on December 2008, was a coordinated international

### pp073117-background-paper-collaboration-between-regional-financing-arrangements-and-the-imf - 22. The SBA for Latvia, approved on December 2008, was a coordinated international

### Latvia SBA (December 2008): objectives, strategy, and outcomes
- Main objectives of the program:
  - (i) arrest the loss of reserves and bank deposits;
  - (ii) restore confidence in the banking sector,
  - (iii) rebuild competitiveness through income policies and structural reforms; and
  - (iv) maintain the fixed exchange rate regime and fulfill Maastricht criteria fiscal targets for euro adoption.
- Macroeconomic adjustment:
  - Characterized by internal devaluation through a combination of wage restraint and productivity gains, which entailed large output costs (Figure 5).
  - The program was an early example of EU-IMF co-financing and is unique with respect to its exit strategy (euro adoption).
- Exchange rate peg:
  - Authorities maintained an unwavering commitment to the exchange rate peg, underpinned by support from the EU (RFA).
  - Fund staff had significant concerns over sustainability of the peg and the authorities’ ability to implement significant fiscal adjustment and income policies.
  - The Latvian lats joined ERM II in 2005. Like other ERM II countries, Latvia was allowed a fluctuation of its currency of ± 15 percent around the central rate vis-à-vis the euro, but it unilaterally maintained a 1 percent fluctuation band.
- Decision to retain the peg:
  - After extensive technical and high-level discussions, the peg was agreed to be the most viable strategy to help restore confidence—based on Latvia’s commitment to the peg, the EC’s focus on minimizing regional spillovers, and authorities’ commitment to undertake any necessary measures to maintain the peg.
  - Maintaining the peg was seen as important to allay concerns over potential regional spillovers, and the adverse impact any devaluation may have on bank balance sheets.
- Outcomes illustrated in Figure 5:
  - GDP and GDP per capita (Real GDP, 2007Q3 = 100): “GDP fell by more than 20 percent from its pre-crisis peak...”
  - Productivity and Wages (Index 2008Q4=100): “...while unit labor costs fell sharply through a combination of wage restraint and labor productivity growth.”
  - Figure axes/indexing preserved as described: GDP Index, Per-capita GDP Index; Labor Productivity, Real Wages, ULC.

### Modalities of cooperation, frictions, and lessons from Latvia
- Modalities evolved over time and were initially informal:
  - No initial formal arrangements for cooperation, (confidential) information sharing, or dispute resolution—informal arrangements established over time.
  - Drafts of program documents (the authorities’ Letter of Intent, Memorandum of Economic and Financial Policies, and Memorandum of Understanding with the EC) were consistently exchanged and cross-checked by program partners.
  - Each institution took funding decisions independently, in line with its own statutes and internal procedures.
  - Initially the EC lacked experience in providing BoP assistance and expertise in program design; Fund team led the macroeconomic framework and program modalities early on. Over time the EC adopted many of the Fund’s practices (policy notes, detailed forecasts, resident representatives).
- Uncoordinated review completion and inconsistent messaging:
  - Tensions arose in summer 2009 over fundamental aspects of program strategy; the Fund was hesitant to conclude the first review due to doubts about sustainability of fiscal targets, while the EU concluded its review and released its second tranche.
  - Uncoordinated completion of reviews weakened the Fund’s program position and created inconsistency between Fund and EC programs. Fund review documents included two alternative scenarios, creating uncertainty and weakening confidence in the overall program.
- Lessons for cooperation (Latvia):
  - Strong ownership by the authorities can serve as an anchor facilitating Fund-RFA cooperation.
  - The RFA can have different priorities, and an early recognition of these differences can facilitate a timely resolution.
  - Uncoordinated decisions can undermine the parties, result in conflicting messages and erosion of credibility, and put the program at risk.

### Romania: repeated collaboration and evolving cooperation (2009, 2011, 2013 SBA)
- Financing and programs:
  - 2009: Agreement to provide Romania with multilateral financing assistance for €20 billion:
    - €5 billion provided by EU’s BoP facility,
    - SDR11.4 billion (around €12.95 billion, exceptional access) by the IMF,
    - Remaining amount financed by other multilateral institutions.
  - 2011: Follow-up financial assistance for €3.5billion (SDR 3.1 billion) with EU providing €1.4 billion (SDR1.2 billion); other multilaterals provided additional funds.
  - 2013: Third assistance program—EU provided €2 billion, in parallel with a Fund SBA for SDR 1.75 billion (around €2 billion).
- Early engagement and division of labor:
  - Early engagement with EU and other lenders was fundamental for a strong 2009 program: large financing, flexible program design, and prioritization were key.
  - Cooperation contributed to financing package and burden sharing; authorities’ commitment to EU targets under the convergence program and the EDP framework helped anchor fiscal consolidation.
  - 2011 SBA: effective division of labor with EC focused on structural reforms (energy, transportation, health); set precedent for future joint support programs.
- Coordination challenges in 2013:
  - 2013 SBA request driven by external vulnerabilities from euro area uncertainties and capital flows to emerging markets.
  - Coordination worked relatively well but could have been enhanced, especially where the Fund had limited experience (e.g., energy and transportation reforms).
  - Fund had to rely on EC and World Bank expertise for key reform elements, complicated by different views on program priorities across institutions.
  - Differences compounded by divergent decision-making procedures, perceptions of insufficient consultation, and inconsistent information sharing—institutions increasingly proceeded on separate tracks.
- Lessons for cooperation (Romania):
  - Parsimonious and focused conditionality is necessary for program success.
  - Divergent institutional decision-making procedures, insufficient consultation, and inadequate information sharing can make agreement protracted and place the program at risk.
  - A well-designed division of labor can facilitate program design and implementation, but when partners have different views on key issues, the division of labor can hinder progress.

### Cross-country contagion and the network contagion model (chapter overview)
- Purpose:
  - Model illustrates contagion across economies arising from large BoP shortfalls emanating from globally or regionally systemic economies, and effectiveness of different layers of the global financial safety net (GFSN) in limiting contagion.
- Contagion mechanism:
  - Initial external shock: suspension of external liability payments by a systemic economy → reduced capital and trade inflows to other countries → decline in gross foreign assets/reserves → external implications for partners.
  - The Fund and RFAs can mitigate contagion through liquidity support.
  - Simulations show: absent stigma, Fund support is generally more effective than RFAs due to global membership and larger resources; contagion can spread more rapidly if Fund involvement is delayed (e.g., due to stigma); collaboration between Fund and RFAs is potentially most effective, especially if RFA hastens Fund involvement.

### A simple model of global interconnectedness and cascading contagion (model equations and mechanics)
- Network setup:
  - Countries interconnected through bilateral asset and trade positions, measured in U.S. dollars.
  - Gross international reserves R increase with current account surplus and inflows from foreign assets, decline with payments on foreign liabilities.
  - Starting from BoP equilibrium, changes in reserves are a function of asset positions and exchange rate changes:
    - ∆푅푖,푡 = ∑ 푎푖푗 푟푗,푡 ∆푒푗,푡 ∈ 푁퐷푡 푗≠푖 ⏟ 푅푒푡푢푟푛 표푛 푛푒𝑡 푓표푟푒𝑖𝑔𝑛 퐴푠푠푒𝑡𝑠 − ∑ 푎𝑗𝑖 푟𝑖,푡 ∆푒𝑖,푡 ⏟ 푅푒푡푢𝑟푛 표푛 푛푒𝑡 푓표푟푒𝑖𝑔𝑛 푙𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠 + ∑푇퐵푖푗 (∆푒푗,푡−1) 푗≠푖 ⏟ 푇푟푎𝑑𝑒 인플ows
  - Notation: 푎푖푗 denotes country i’s asset holding against country j, 푟푗,푡 is country j’s specific interest rate on its liabilities, and ∆푒푖,푡−1∗ is the nominal effective exchange rate of the previous period.
- Liquidity and default:
  - If reserves sufficient (푅푖>0) all external liabilities honored, country ∈ 푁퐷푡 (non-defaulted set).
  - If reserves depleted (푅푖=0), country cannot make foreign payments → external default → i ∈ 퐷푡 (defaulted set).
- Contagion trigger:
  - Initial suspension of external payments by country h (assumed systemic): country h stops making payments 푎푖ℎ,푡 푟ℎ,푡.
  - For exposure to h, initial loss in international reserves for any country i: ∆푹푰,푡 = −푎푖푕,푡 푟푕푡 푒푕,푡. If reserves depleted, the country also defaults.
- Cascading default condition:
  - ∀ 푖 ∈푁퐹푡 if ∑푟 aℎ𝑖 ℎ∈퐹𝑖 > 푅𝑖 → Country i defaults: i ∈ F_{t+1}.

### Role of the GFSN, timing, and access limits
- Liquidity support via GFSN:
  - Countries can draw on credit lines from the Fund (퐸퐹퐼푀퐹) or from RFAs (퐸퐹푅퐹퐴) when BoP needs arise.
  - Drawdowns support reserve accumulation up to limits determined by maximum access in Fund and RFA:
    - ∆푅푖 = ∑ 푎푖푗 푟푗,푡 푒푗,푡 ∈ 푁퐹푡 푗≠푖 ⏟ 푅푒푡푢𝑟푛 표푛 푛푒𝑡 푓표푟푒𝑖𝑔𝑛 퐴푠푠푒𝑡𝑠 − ∑ 푎𝑗𝑖 푟𝑖,푡 푒𝑖,푡 ⏟ 푅푒푡푢𝑟푛 표푛 푛푒𝑡 푓표푟푒𝑖𝑔𝑛 푙𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠 + ∑푇퐵푖푗 (푒푗푖,푡−1∗) 푗≠푖 ⏟ 푇푟푎𝑑𝑒 푟𝑒𝑣𝑒𝑛𝑢𝑒𝑠 + 퐸퐹𝑖𝑡 ⏟ 𝐺𝐹𝑆𝑁 𝑓𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔
  - 퐸퐹푖𝑡 = 퐸퐹푅퐹퐴; 푖,푡 + 퐸퐹퐼푀퐹; 푖,푡
  - External financing cannot exceed maximum access: 퐸퐹푅퐹퐴; 푖,푡 ≤ 퐸퐹̅̅̅̅푅퐹퐴;푖,푡 and 퐸퐹퐼푀퐹; 푖,푡 ≤ 퐸퐹̅̅̅̅퐼푀퐹;푖,푡.
- Timing and stigma:
  - Model allows delays in access to IMF support (stigma): 퐸퐹퐼푀퐹; 푖,푡 = 0 for as long as Fund involvement is delayed.
  - RFAs cover fewer countries but their assistance likely materializes more rapidly, partly due to specific liquidity instruments without structural components and simpler processes.

### Empirical implementation and data
- Dataset and calibration:
  - Model applied to dataset covering 62 countries representing 78 percent of global GDP.
  - Bilateral asset positions and trade flows set to match actual exposures as of 2016.
  - Asset data composed of interbank liabilities (BIS locational statistics) and portfolio investment (CPIS).
  - Trade flows from Direction of Trade Statistics (DOTS).
  - Model calibrated to quarterly frequency starting with situation equivalent to the first quarter of 2017.
- RFAs considered:
  - Chiang Mai Initiative Mechanism (CMIM), BRICS (Brazil, Russia, India, China and South Africa), the Latin American Reserve Fund (FLAR), and the European Stability Mechanism (ESM).

*International Monetary Fund — pp073117-background-paper-collaboration-between-regional-financing-arrangements-and-the-imf (selected excerpts)*

### 38. For this illustrative exercise, the size of the initial shocks is calibrated to simulate to

### 38. For this illustrative exercise, the size of the initial shocks is calibrated to simulate to

### D. Results
- Purpose of simulations
  - Illustrate impact of global and regional systemic shocks and the effectiveness of Global Financial Safety Net (GFSN) constellations.
  - Global shock: a simultaneous shock from different globally relevant countries.
  - Regional shock: emanates from a single regional power.
  - In each case, the shock leads to a permanent suspension of interest payments to all other countries in the sample.
- Iteration/convergence procedure
  - Generally, convergence is reached when for two periods, no additional default occurs.
  - Results presented use six iterations for comparability across GFSN constellations.
  - Balance of payments is assumed in equilibrium; exchange- and interest rates are normalized. Any drawdown of reserves is exclusively a function of changes since the starting situation.
- Key simulation findings
  - If countries rely exclusively on their own means (their own reserves), roughly 20 countries would be affected by contagion following the shocks.
  - Immediate access to Fund resources reduces the number of affected countries to 9 in the case of a globally systemic shock—reflecting the central role of the IMF in the GFSN.
  - Collaboration between the IMF and RFAs can halve the number of contagion events compared with Fund-only or RFA-only responses.
- Table 6 — The Effectiveness of the GFSN (Number of countries affected by the initial shock)
  - Columns (in order): IMF with no lag | IMF with 2-quarter lag | RFA only | IMF (no lag) + RFA | IMF (2-quarter lag) + RFA
  - Global shock: 9 | 19 | 13 | 5 | 12
  - Regional shock: 0 | 5 | 7 | 0 | 2
- Role of RFAs
  - RFAs in isolation can play a significant role, mostly when dealing with regional shocks.
  - RFA support tends to be more effective when the initial shock is regional, provided the shock is not synchronized across all RFA members or does not significantly affect the largest members of the RFA.
  - Simulations show RFAs can reduce the number of contagion events to 13 countries if the initial shock is global, and reduce contagion by more than half if the shock is regional.
- Importance of speed of IMF intervention
  - The Fund’s effectiveness in preventing contagion hinges on the speed of intervention.
  - Comparing a no-delay scenario with a 2-quarter stigma (delay) scenario: with stigma, contagion cascades materialize rapidly, undermining the effectiveness of the Fund when acting alone and reducing its marginal contribution when collaborating with RFAs.

### Robustness analysis
- Qualitative robustness
  - Conclusions are robust across different parameter values in the model.
  - The relative performance of different constellations of the GFSN remains unchanged under robustness checks.
- Sensitivity to initial shock size
  - The number of defaults depends crucially on the assumed size of the initial default.
  - The initial shock is set equal to the implied quarterly interest payments due, based on total outstanding liabilities and an appropriate interest rates, but excluding principal repayment for this illustrative exercise.
  - The number of defaults may increase more than proportionally with larger shocks.
  - As credit lines with international financial institutions shrink relative to the size of the initial shock, their usefulness is quickly reduced; a single additional country falling under contagion can, via multiple interactions, quickly increase stress on the rest of the countries.
  - Ultimately, the number of potential countries falling under contagion would increase in a non-linearly manner, though the relative performance ranking of GFSN constellations remains robust.

*Source: Fund staff calculations, pp073117-background-paper-collaboration-between-regional-financing-arrangements-and-the-imf*

---


_Source: https://www.imf.org/-/media/files/publications/pp/2017/pp073117-background-paper-collaboration-between-regional-financing-arrangements-and-the-imf.pdf_