## _cr12134

## Source details

**Canonical URL:** [_cr12134](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2012/_cr12134.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2012/_cr12134.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2012/_cr12134.pdf.json)

---

### I. DEBT MARKET DEVELOPMENT — Overview and Key Findings
- Purpose: Assess Macedonia’s public debt markets and present recommendations for further development.
- Rationale:
  - Domestic market development will diversify funding sources and reduce reliance on volatile external financing.
  - Increased liquidity and depth will allow better management of risks.
  - Establishing a reference yield curve will generate positive externalities for the private sector.
  - Market development requires a medium-term comprehensive strategy and many incremental steps.
- Current status (summary):
  - Domestic market started in 2004; first Eurobond in 2005.
  - Domestic market remains shallow, with some 5 percent of GDP in regular bills and bonds outstanding, almost all at short maturities.
  - Private external financing represents an additional 6 percent of GDP.
- Data note: Debt statistics are based on data through 2011 Q4.

### II. Market Evolution — Historical developments and crisis impacts
- Timeline highlights:
  - 2004: First domestic treasury bills for regular budget financing issued.
  - 2005: First Eurobond issued; issuance extended to 2, 3 and 5-year treasury bonds by 2007.
  - 2009–2010: Crisis prompted concentration on 3- and 6–month segments, switch from interest rate to volume tenders, and introduction of 3- and 6-month euro-linked, denar-denominated securities.
  - December 2011: Government procured a 5-year €130 million syndicated loan supported by a Policy-Based Guarantee from the World Bank.
- Crisis effects and policy response:
  - Average maturity of outstanding debt shortened considerably.
  - Spread between local currency and FX-indexed debt diminished from some 400 basis points to just 10 bps.
  - From end-2010, interest rate on local currency T-bills set at a spread of 20–30 bps over the central bank’s 30-day policy rate.
- Pre-crisis market-building practices:
  - Three-year rolling medium term debt strategy, regular market reports, predictable auction calendar.
  - Some treasury bills issued to mop up excess liquidity in the banking system.

### III. Current Market Structure and Investor Base (end-2011)
- General government debt:
  - Total: some €2 billion or 28 percent of GDP.
  - Almost three quarters is external.
- External debt:
  - Total external debt: €1.6 billion.
  - Close to 60 percent from multilateral creditors.
  - Two outstanding Eurobonds represent about 21 percent of general government external debt.
- Domestic government securities:
  - Total: some €507 million equivalent.
    - €355 million in regular bills and bonds (“marketable securities”).
    - €150 million structural bonds (transition-related compensation), almost all fixed rate.
  - Within marketable securities, short-term treasury bills make up some 95 percent; remaining 5 percent are recently issued 5-year treasury bonds.
  - Almost 50 percent of domestic government securities is indexed to or denominated in euros.
- Ownership structure (November 2011):
  - Banks hold almost two-thirds (about 65 percent) of regular government securities.
  - “Other” (Deposit Insurance Fund, private companies, individuals) is the second largest category.
  - Pension funds and life insurance companies:
    - Second-pillar pension funds asset base: some €250 million at end-2011.
    - Insurance sector total assets: some €210 million; holds about €15 million in domestic government securities.
    - Life insurance assets: less than €30 million and growing at a 15 percent annual rate; life companies invest some 40 percent of assets in government securities (both domestic and Eurobonds).
  - Foreign investors hold no domestic government debt despite absence of regulatory barriers.

### IV. Regional Comparison — Debt levels and market structure
- Government debt trends:
  - Between 2000 and 2010, government debt halved from over one-half to close to one-quarter of GDP.
  - Macedonia had lower government debt as a share of GDP in 2010 than in 2006, in contrast with most regional peers.
- Domestic market size and maturity:
  - End-2010: Macedonia’s domestic debt just under 8 percent of GDP, among the lowest in the region.
  - Maturity structure concentrated on the short end; short-term debt constitutes almost all domestic marketable debt.
  - Most regional peers feature short-term debt of less than half of total and many have increased average maturity over time.
- Debt strategy and benchmarks:
  - Debt management strategy covers 3 years, comparable to peers.
  - On operational risk benchmarks, Macedonia covers only currency composition; most regional countries cover maturity, variability and duration.
  - Since 2009 Macedonia has not published an ex-post evaluation of its strategy.

### V. Going Forward: Market Development and Risk Mitigation — Recommendations
- Main recommendations (Box 1):
  - Increase the size of the domestic government debt market to reduce reliance on volatile external funding.
  - Lengthen domestic maturities, and be prepared to pay market interest rates.
  - Enhance liquidity through fungibility and attention to secondary market development.
  - Measure risks explicitly and benchmark them against a chosen debt profile.
  - Reduce refinancing risk through pre-financing and avoiding bunching of maturities.
  - Further develop institutions and practices to achieve these goals.
- Specific market-stimulating measures:
  - Use interest rate tenders to provide price discovery and meet cash management needs (current practice of setting rates and allowing volume to adjust frequently results in undersubscribed auctions).
  - Develop the secondary market by facilitating repo market development; concentrating issuance on a limited number of large benchmark issues; reopening outstanding issues when they reach the next benchmark maturity rather than issuing new securities.
  - Predictable issuance via a regular calendar of issuance to create stable demand.
  - April 2012 NBRM repurchase facility: banks can obtain liquidity against holdings of Treasury and central bank securities, which should increase demand by reducing illiquidity risk.
- Market capacity and liquidity indicators:
  - Stock of central bank bills: some MKD 33 billion, over 7 percent of GDP, indicating excess liquidity and room to increase domestic debt issuance without crowding out private lending.
  - Increased Treasury issuance could reduce the need for central bank bills and relieve the NBRM’s balance sheet of some of the cost of absorbing liquidity.
- Example noted: Hungary — development through five building blocks: debt management, investor base, market infrastructure, regulatory environment, and monetary policy and operations.

### Risk Management and Medium-Term Debt Strategy
- Strategy design principles:
  - Set risk benchmarks; explicitly capture tradeoffs between costs and risks; take a medium-term perspective.
  - Model refinancing risks, interest rate risks, currency risks on current and alternative public debt portfolio compositions.
- Stress tests:
  - Perform stress tests including severe, but credible shocks.
  - Base shocks on historical experience from Macedonia or similar countries, or use discretionary shocks.
  - Judge outcomes against risk parameters in the debt strategy.

### Refinancing Risk — Assessment and Mitigation
- Refinancing risk is significant:
  - Large Eurobonds maturing in 2013 and 2015 must be repaid.
  - PLL purchase due in 2014–16.
  - Bank loan guaranteed by the World Bank PBG due in 2016.
  - Domestic debt average maturity: just 6 months (implying need to roll over existing debt on average 2 times per year).
- Past market volatility examples:
  - 2009: government issued a 3½-year Eurobond at a yield of 9⅞ percent (more than double the 4⅝ percent yield of the 10-year issuance in 2005).
  - 2010: a planned Eurobond issuance was canceled after market rates spiked in the run-up to the Greek program.
- Mitigation measures:
  - Pre-finance Eurobond rollovers and lengthen domestic maturities.
  - Authorities’ April 2012 announcement of a €250 million loan agreement with a foreign bank noted as pre-financing the €175 million Eurobond maturing in January 2013.
  - Initiated quarterly auction of 5-year bonds starting in September 2011 to lengthen domestic maturities.

### Currency Risk and De-euroization of Debt
- Currency exposure:
  - All external debt is in FX.
  - Almost half of outstanding domestic government debt is FX-linked.
  - Taken together, almost 90 percent of general government debt is denominated in or linked to foreign currency.
- Assessment:
  - High exposure multiplied by low probability of a shock may make currency risk appear low given peg stability over past 15 years, but reducing it would add policy flexibility in unforeseen shocks.
- Recommended actions:
  - Gradually increase the share of domestic denar-denominated, non FX-linked debt.
  - Increase the share of domestic debt in fiscal financing.
- Expected benefits:
  - Develop the domestic currency yield curve.
  - Improve monetary policy transmission mechanism.
  - Promote greater use of denars in financial transactions.

### Institutional Arrangements and Capacity Needs
- Key gaps and needs:
  - Capacity enhancement for the Ministry of Finance Public Debt Management Department (PDMD) is crucial.
  - PDMD was severely understaffed and without a department head for almost 2 years.
  - Under a recent Ministry decision, the department was merged with the department for EU harmonization and international financial institutions, and a new department head, together with three additional staff, were appointed.
- Additional recommendations:
  - Strengthen implementation of Public Debt Law responsibilities despite staffing gaps.
  - Improve communications with investors via proactive practices (publication of regular reports; constant contacts with investors, regulators, financial press, and rating agencies); assign front or middle office staff to these tasks.
  - Formalize information exchange with the NBRM; consider setting up a high-level debt management committee to coordinate strategy and oversee implementation.
  - Establish an operational sub-committee to discuss granular technical issues regarding issuance coordination.
  - Liquidity committee (meets weekly) could improve functioning if it works against a clear definition of the optimal level of cash balances.
- Expected outcomes from a medium-term strategy:
  - Minimize medium-term costs subject to keeping risks within appropriate pre-set bounds.
  - Benefits include lower interest expense and lower risks over the medium term and positive externalities for the real economy through a denar yield curve usable in pricing private sector loans and bonds.

### Euroization, Prudential Framework, and Monetary Policy
- Causes and dynamics of euroization:
  - Driven by inflation volatility (current and memories of past hyperinflation), low exchange rate volatility and potential implicit guarantees under the exchange rate peg, and increased demand and supply of euros from regional integration.
  - Borrower myopia unlikely main cause; high deposit euroization despite higher returns on local currency deposits suggests residents understand currency risk.
- Prudential and regulatory framework (Box 1):
  - Reserve requirements: 10 percent on denar deposits, 13 percent on FX deposits, and 20 percent on FX-indexed deposits.
  - Net open FX position limit: 30 per cent of the banks’ own funds (includes FX and FX-indexed assets and liabilities).
  - Capital adequacy: additional capital requirements for currency risk calculated by multiplying the sum of the absolute amount of the net position in gold and the bank’s aggregate FX position, by 8%.
  - Identification and monitoring of unhedged FX borrowers: banks should monitor borrower FX exposures, stress test cash flows against depreciation at least annually, determine adequate collateral, set quantitative limits on FX or FX-indexed loans, and perform portfolio stress tests when FX and FX-indexed loans to residents exceed 15 percent of total credit to residents net of specified offsets.
  - Hedging definition: client considered hedged if expected FX inflows, FX and FX-indexed claims cover at least 80 percent of total expected FX outflows, FX and FX-indexed liabilities.
  - Documentation: credit file must contain all relevant data for assessing identity, creditworthiness, regularity of settling liabilities and collateral quality.
- Benefits and limits of de-euroization under the peg:
  - Primary benefit: reducing balance sheet risks and allowing exchange rate flexibility to serve as a “safety valve” in tail risk scenarios.
  - Limited gains for monetary policy transmission under the peg; peg has been the main policy anchor.
- Cross-country lessons:
  - Successful de-euroization cases tended to have flexible exchange rate regimes and were supported by macro stability, active public debt management, and developed domestic financial markets.
  - Prudential measures that can support de-euroization: higher liquidity requirements on FX deposits or FX loans, higher capital and provisioning requirements on FX loans to unhedged borrowers, lower loan-to-value ratios on FX loans.
- Conclusions and staff recommendations:
  - Euroization is persistent; Macedonia likely to remain euroized for a long time.
  - Main recommendation: provide incentives and a supportive macro-financial context for gradual reduction of euroization through prudential policies, maintaining macro stability, and developing domestic debt markets.
  - Specific actions: maintain prudent macro policies and macro stability; improve public debt management and rely more on longer term denar loans to de-euroize the public sector balance sheet and lengthen the denar yield curve; develop a deep and liquid domestic capital market to provide investment opportunities in denar.

### Crises, Deposits, and Monetary Policy Responses
- Deposit dynamics in crises:
  - Private sector deposits dropped by 14 percent in the security crisis of 2001 and by 8 percent in 2008–9.
  - Corporate deposits fell more steeply and persistently than household deposits.
  - By the end of both crisis periods the degree of household euroization had increased significantly.
- Selected episode data (table excerpts):
  - Feb 01 – Sep 01: 8 | 253.0 | 24.3 | 6.4 | 4.8
  - May 02 – Mar 03: 11 | 232.7 | 28.2 | 4.6 | 3.4
  - Oct 03 – Nov 04: 14 | 9.0 | 1.0 | 3.8 | 3.2
  - Nov 07 – Apr 08: 6 | 95.8 | 4.3 | 4.1 | 3.9
  - Oct 08 – May 09: 8 | 732.5 | 30.1 | 4.3 | 3.2
  - (Table header context: "Months of imports Cumulative loss in Reserves (mn dols) (percent*) at start at end"; source: NBRM. "* of FA on month prior to crisis")
- Monetary policy and defending the peg:
  - NBRM tightened policy rate when reserves were “low and falling,” defending the peg.
  - Examples:
    - January 2001 privatization proceeds kept foreign assets comfortable after a 24.3% loss in reserves; NBRM raised policy rate to 20 percent.
    - 2003–4: minor reserve drop (1%) but import coverage fell to 3.2 months; NBRM raised policy rate and agreed IMF program relying on privatization proceeds to strengthen peg.
    - First half of 2008: central bank tightened policy despite relatively stable reserves to counter overheating (high inflation, high credit growth, current account deterioration).
- Complementary crisis measures:
  - 2001: increased reserve requirements; temporary restriction on FX sales to big firms and bureaus; shortened export proceeds surrender period from 180 to 30 days.
  - 2003–4: introduced reserve requirements on FX deposits; shifted CB-bills auctions to volume tender.
  - 2008: increased capital adequacy requirements for overdraft and credit card loans; broadened base for reserve requirements.
  - 2009: introduced liquidity ratios requirements; increased reserve requirements for FX and FX indexed deposits.

### Model of Semi-Autonomous Monetary Policy — Stylized Results and Policy Implications
- Model structure and assumptions:
  - Central bank dual objective: defend the peg and influence inflation and output gap when exchange rate not under pressure.
  - Monetary autonomy assumed: domestic interest rates can differ from UIP.
  - New-Keynesian dynamics for inflation (π_t) and output gap (x_t); Taylor-rule behavior in absence of exchange rate pressure.
  - Reserve dynamics: Δr_t driven by deviations from UIP; reserves r_t must be kept above threshold r̄.
  - Central bank minimizes loss combining deviations of policy rate from Taylor rule and a reserve-related cost; reduced-form analysis assumes myopic behavior.
- Stylized calibration and implications:
  - Weight on Taylor-rule component becomes zero if reserves are 30 percent below the reserve target (calibration based on NBRM reactions to ~30 percent reserve drops).
  - When reserves below threshold (θ<1) policy rate response to inflation or output shocks is muted (capping from below or above).
  - Result: lower volatility of interest rates compared to unconstrained Taylor-rule case.
- Policy conclusions:
  - NBRM should continue timely monetary responses based on monitoring reserve adequacy, balance of payments dynamics, credit growth, trends in euro-denominated to denar deposits, and Eurozone interest rates.
  - Preserve confidence in the peg by holding adequate international reserves; accept limited policy rate volatility under the peg.

### Key numeric facts and thresholds cited
- 9.9 percent of total FX and FX-indexed loans (context: referenced in document header).
- Reserve requirements: 10 percent (denar deposits), 13 percent (FX deposits), 20 percent (FX-indexed deposits).
- Net open FX position limit: 30 per cent of the banks’ own funds.
- Capital charge multiplier for currency risk: 8%.
- Stress-test trigger for annual portfolio assessment: FX and FX-indexed loans to residents exceed 15 percent of total credit to residents (net of specified offsets).
- Hedged FX position definition threshold: 80 percent coverage of expected FX outflows by expected FX inflows and claims.
- De-euroization success criterion (Reinhart, Rogoff, and Savastano (2003)): deposit euroization falls by at least 20 percentage points and remains below 20 percent.
- Peg history: pegged to the euro since January 2002; previously pegged to the Deutsch Mark since 1995.
- Historical stress periods: drops in foreign assets averaged 17 percent and lasted between 6 and 14 months.
- Domestic market totals and shares repeatedly cited:
  - Domestic market some 5 percent of GDP in regular bills and bonds outstanding.
  - Private external financing represents an additional 6 percent of GDP.
  - General government total debt: some €2 billion or 28 percent of GDP.
  - Total external debt: €1.6 billion.
  - Domestic government securities total: some €507 million equivalent (€355 million marketable; €150 million structural).
  - Central bank bills stock: some MKD 33 billion, over 7 percent of GDP.
  - Second-pillar pension funds asset base: some €250 million at end-2011.
  - Insurance sector total assets: some €210 million; holds about €15 million in domestic government securities.
  - Life insurance assets: less than €30 million and growing at a 15 percent annual rate; life companies invest some 40 percent of assets in government securities.

*IMF staff report excerpt (FYR Macedonia).*

### References .............................................................................................................

### _cr12134 - References .............................................................................................................

### I. DEBT MARKET DEVELOPMENT IN FYR MACEDONIA — Overview and Key Findings
- Purpose: Assess Macedonia’s public debt markets and present recommendations for further development.
- Rationale:
  - Domestic market development will diversify funding sources and reduce reliance on volatile external financing.
  - Increased liquidity and depth will allow better management of risks.
  - Establishing a reference yield curve will generate positive externalities for the private sector.
  - Market development requires a medium-term comprehensive strategy and many incremental steps.
- Current status (summary):
  - Domestic market started in 2004; first Eurobond in 2005.
  - Domestic market remains shallow, with some 5 percent of GDP in regular bills and bonds outstanding, almost all at short maturities.
  - Private external financing represents an additional 6 percent of GDP.
- Data note: Debt statistics are based on data through 2011 Q4.

### II. Market Evolution — Historical developments and impacts of the crisis
- Timeline highlights:
  - 2004: First domestic treasury bills for regular budget financing issued.
  - 2005: First Eurobond issued; issuance extended beyond 3-month T-bills to include 2, 3 and 5-year treasury bonds by 2007.
  - 2009–2010: Global economic crisis prompted concentration on 3- and 6–month segments, switch from interest rate to volume tenders, and introduction of 3- and 6-month euro-linked, denar-denominated securities.
  - December 2011: Government procured a 5-year €130 million syndicated loan supported by a Policy-Based Guarantee from the World Bank.
- Effects of crisis and policy response:
  - Average maturity of outstanding debt shortened considerably.
  - Spread between local currency and FX-indexed debt diminished from some 400 basis points to just 10 bps.
  - From end-2010, interest rate on local currency T-bills set at a spread of 20–30 bps over the central bank’s 30-day policy rate.
- Market-building practices used pre-crisis:
  - Three-year rolling medium term debt strategy, regular market reports, predictable auction calendar.
  - Some treasury bills issued specifically to mop up excess liquidity in the banking system.

### III. Current Market Structure and Investor Base — Composition and ownership (end-2011)
- General government debt:
  - Total: some €2 billion or 28 percent of GDP.
  - Almost three quarters is external.
- External debt:
  - Total external debt: €1.6 billion.
  - Close to 60 percent from multilateral creditors.
  - Two outstanding Eurobonds represent about 21 percent of general government external debt.
- Domestic government securities:
  - Total: some €507 million equivalent.
    - €355 million in regular bills and bonds (“marketable securities”).
    - €150 million structural bonds (transition-related compensation), almost all fixed rate.
  - Within marketable securities, short-term treasury bills make up some 95 percent; remaining 5 percent are recently issued 5-year treasury bonds.
  - Almost 50 percent of domestic government securities is indexed to or denominated in euros.
- Ownership structure (data of November 2011):
  - Banks hold almost two-thirds (about 65 percent) of regular government securities.
  - “Other” (Deposit Insurance Fund, private companies, individuals) is the second largest category.
  - Pension funds and life insurance companies currently small investors but growing:
    - Second-pillar pension funds asset base: some €250 million at end-2011.
    - Insurance sector total assets: some €210 million; holds about €15 million in domestic government securities.
    - Life insurance assets: less than €30 million and growing at a 15 percent annual rate; life companies invest some 40 percent of assets in government securities (both domestic and Eurobonds).
  - Foreign investors hold no domestic government debt despite absence of regulatory barriers.

### IV. Regional Comparison — Debt levels and market structure
- Government debt trends:
  - Between 2000 and 2010, government debt halved from over one-half to close to one-quarter of GDP.
  - In contrast with most regional peers, Macedonia had lower government debt as a share of GDP in 2010 than in 2006.
- Domestic market size and maturity:
  - End-2010: Macedonia’s domestic debt just under 8 percent of GDP, among the lowest in the region.
  - Maturity structure concentrated on the short end; short-term debt constitutes almost all domestic marketable debt.
  - Most regional peers feature short-term debt of less than half of total and many have increased average maturity over time.
- Debt strategy:
  - Macedonia’s debt management strategy covers a period of 3 years, comparable to peers.
  - On operational risk benchmarks, Macedonia covers only currency composition, while most countries in the region cover maturity, variability and duration.
  - Since 2009 Macedonia has not published an ex-post evaluation of its strategy.

### V. Going Forward: Further Market Development and Risk Mitigation — Recommendations and measures
- Box 1: Main Recommendations on Market Development
  - Increase the size of the domestic government debt market to reduce reliance on volatile external funding.
  - Lengthen domestic maturities, and be prepared to pay market interest rates.
  - Enhance liquidity through fungibility and attention to secondary market development.
  - Measure risks explicitly and benchmark them against a chosen debt profile.
  - Reduce refinancing risk through pre-financing and avoiding bunching of maturities.
  - Further develop institutions and practices to achieve these goals.
- Specific measures to stimulate market development:
  - Use interest rate tenders to provide price discovery and meet cash management needs (current practice of setting rates and allowing volume to adjust frequently results in undersubscribed auctions).
  - Develop the secondary market by:
    - Facilitating repo market development.
    - Concentrating issuance on a limited number of large benchmark issues.
    - Reopening outstanding issues when they reach the next benchmark maturity rather than issuing new securities.
  - Predictable issuance via a regular calendar of issuance to create stable demand.
  - The April 2012 NBRM repurchase facility: banks can obtain liquidity against holdings of Treasury and central bank securities, which should increase demand by reducing illiquidity risk.
- Market capacity and liquidity indicators:
  - Stock of central bank bills: some MKD 33 billion, over 7 percent of GDP, indicating excess liquidity and room to increase domestic debt issuance without crowding out private lending.
  - Increased Treasury issuance could reduce the need for central bank bills and relieve the NBRM’s balance sheet of some of the cost of absorbing liquidity.
- Example: Hungary (Box 2) — illustrating how a small country can develop a well-functioning government securities market through consistent, integrated market-building measures across five building blocks: debt management, investor base, market infrastructure, regulatory environment, and monetary policy and operations.

*Prepared by Eugen Tereanu and Alexander Tieman; content reflects sections and figures excerpted from the source document.*

### 15.      It will take time to increase the size of the domestic debt market. The size of the

### 15.      It will take time to increase the size of the domestic debt market. The size of the 

### Size of the domestic market and role of external financing
- The size of the domestic market is currently modest in relation to forthcoming fiscal financing needs and (domestic and external) debt rollovers.
- It will take time to increase the domestic debt market, and external debt will remain an important financing source in the short to medium-term.
- Even once the domestic debt market is fully developed, external financing will continue to play a role.

### Risk mitigation and medium-term debt strategy
- Managing risks and analyzing risk-cost trade-offs are an important aspect of a medium-term debt strategy.
- Active management of risks embedded in the debt portfolio should be done in the context of a comprehensive debt strategy that:
  - sets risk benchmarks;
  - explicitly captures tradeoffs between costs and risks;
  - takes a medium-term perspective (acknowledging that minimizing costs over the medium run may not be consistent with cost minimization in any single year), including with respect to market development.
- The debt management strategy should explicitly model:
  - refinancing risks;
  - interest rate risks;
  - currency risks;
  - on the current as well as potential alternative public debt portfolio compositions.
- Performing stress tests of the debt portfolios is recommended to quantify risks. Stress tests should:
  - include severe, but credible shocks;
  - base the magnitude of shocks on historical experience from either Macedonia or other countries in similar circumstances, or be of a more discretionary nature;
  - judge outcomes against the risk parameters laid out in the debt strategy to assess acceptability.

### Refinancing risk for FYR Macedonia
- Refinancing risk is significant:
  - large Eurobonds maturing in 2013 and 2015 must be repaid;
  - PLL purchase due in 2014–16;
  - a bank loan guaranteed by the World Bank PBG due in 2016.
- Market volatility has been high since the onset of the global crisis, due to external rather than domestic factors.
  - 2009: government issued a 3½-year Eurobond at a yield of 9⅞ percent (more than double the 4⅝ percent yield of the 10-year issuance in 2005).
  - 2010: a planned Eurobond issuance was canceled after market rates spiked in the run-up to the Greek program.
- Domestic debt features a very short average maturity of just 6 months, implying the need to roll over existing debt on average 2 times per year.

### Measures to mitigate refinancing risk
- Pre-finance Eurobond rollovers and lengthen domestic maturities:
  - Given gradual deepening of the domestic market, the government will need to issue net new private external debt to meet fiscal financing needs and roll over maturing issues.
  - To decrease associated refinancing risk, the authorities should ensure an adequate buffer of deposits by pre-funding rollovers, instead of waiting to issue shortly before existing maturities.
  - Authorities’ announcement in April 2012 of a €250 million loan agreement with a foreign bank is noted as pre-financing the €175 million Eurobond maturing in January 2013.
  - On the domestic side, issuing longer-term paper would increase average maturity and lower refinancing risk; the authorities initiated a quarterly auction of 5-year bonds starting in September 2011.

### Currency risk and de-euroization of debt
- Currency exposure is considerable:
  - all external debt is in FX;
  - almost half of outstanding domestic government debt is FX-linked (Table 2).
  - Taken together, almost 90 percent of general government debt is denominated in or linked to foreign currency.
- Currency risk (high exposure multiplied by low probability of a shock) may not appear high given the stability of the peg over the past 15 years, but reducing it would add policy flexibility in the event of unforeseen shocks.
- Recommended actions:
  - Gradually increase the share of domestic denar-denominated, non FX-linked debt.
  - Increase the share of domestic debt in fiscal financing.
- Expected benefits of these actions:
  - develop the domestic currency yield curve;
  - improve the monetary policy transmission mechanism;
  - promote greater use of denars in financial transactions.

### Institutional arrangements and capacity needs
- Market development, risk analysis, and mitigation should be supported by strengthening institutional arrangements.
- Key gaps and needs:
  - Capacity enhancement for the Ministry of Finance Public Debt Management Department (PDMD) is crucial.
  - The PDMD was severely understaffed and without a department head for almost 2 years, effectively preventing adequate task performance.
  - Under a recent Ministry decision, the department was merged with the department for EU harmonization and international financial institutions, and a new department head, together with three additional staff, were appointed.
- Additional institutional improvements recommended:
  - Strengthen implementation of Public Debt Law responsibilities despite staffing gaps.
  - Improve communications with investors via more proactive practices (publication of regular reports on the public debt market; constant contacts with investors, regulators, financial press, and rating agencies); assign front or middle office staff to these tasks.
  - Formalize information exchange with the NBRM; consider setting up a high-level debt management committee to coordinate strategy and oversee implementation.
  - Establish an operational sub-committee to discuss granular technical issues regarding issuance coordination.
  - Liquidity committee (meets weekly) could improve functioning if it works against a clear definition of the optimal level of cash balances.

### Expected outcomes from a medium-term debt strategy
- By developing and following a medium-term debt strategy along the lines described:
  - Macedonia can achieve the goal of minimizing medium-term costs subject to keeping risks within appropriate pre-set bounds.
  - Further gradual market development moves, guided by a comprehensive debt strategy, would bring this goal within reach.
  - Benefits would include lower interest expense and lower risks over the medium term.
  - Developing a domestic yield curve would yield positive externalities for the real economy by establishing a denar yield curve usable in pricing private sector loans and bonds.

### Box 3 — Institutional Arrangement for Debt Management (summary)
- Legal and regulatory framework exists, but implementation exhibits gaps mainly due to severe staff shortages.
- PDMD organization follows best practice (front, middle, back office), but:
  - PDMD lacked a department head for about two years;
  - Department remains understaffed;
  - Middle office effectively consists of a single individual, inadequate for comprehensive debt management analyses.
- Communications with investors occur but could be more proactive (regular public reports; ongoing stakeholder contact).
- Relationship between debt management and monetary policy functions well but lacks a formalized information-exchange structure:
  - Proposal: set up a high-level debt management committee to coordinate strategy and oversee implementation;
  - Proposal: establish an operational sub-committee for granular issuance coordination;
  - Liquidity committee meets weekly; could work better with a clear definition of the optimal level of cash balances.
- Source cited within box: IMF, 2011.

*Source: _cr12134 - 15.      It will take time to increase the size of the domestic debt market. The size of the*

### 9.9 percent of total FX and FX-indexed loans. In the tail risk event of a large depreciation,

### _cr12134 - 9.9 percent of total FX and FX-indexed loans. In the tail risk event of a large depreciation,

### Causes and dynamics of euroization in FYR Macedonia
- Loan euroization is driven by:
  - inflation volatility, including both current volatility and memories of past hyperinflation episodes;
  - low exchange rate volatility and potential implicit guarantees perceived under the exchange rate peg;
  - larger demand and supply of euros as Macedonia becomes more integrated with the region in trade and financial flows.
- Borrower behavior:
  - Borrower myopia is unlikely to be a main cause: high deposit euroization despite higher returns on local currency deposits suggests residents understand currency risk.
- Supply-side factors:
  - Deposit euroization is prevalent and contributes to high loan euroization.
  - Foreign currency funding from abroad is limited, but deposit euroization is among the highest in the region.
  - FX deposits appear very sensitive to residents’ perceptions of macroeconomic risk.

### Prudential and regulatory framework (Box 1: Regulation of FX-Induced Credit Risk)
- Differential reserve requirements:
  - Reserve requirements are 10 percent on denar deposits, 13 percent on FX deposits, and 20 percent on FX-indexed deposits.
- Net open FX position:
  - A bank’s aggregate net open FX position at the end of any business day should not exceed 30 per cent of the banks’ own funds. The FX position includes both FX and FX-indexed assets and liabilities.
- Capital adequacy:
  - The additional capital requirements for currency risk are calculated by multiplying the sum of the absolute amount of the net position in gold and the bank’s aggregate FX position, by 8%.
- Identification and monitoring of unhedged FX borrowers:
  - Banks should monitor and assess the borrower’s FX claims and liabilities structure by currency and maturity.
  - Banks should stress test borrower cash flows against a depreciation of denar against the currency in which the loan is denominated, at least on an annual basis.
  - Banks should determine adequate collateral considering the borrower’s capacity to repay after a depreciation.
  - Banks should determine quantitative limits on the amount of each FX or FX-indexed loan to residents and on the share of FX and/or FX-indexed credit exposure of residents in the bank’s total credit exposure.
  - Banks should assess exchange rate fluctuation influence on loan portfolios by stress tests or other simulations at least once a year if the bank’s FX and FX-indexed loans to residents exceed 15 percent of its total credit to residents, net of the bank’s FX deposits and the FX or FX-indexed investments of securities issued by the Macedonian government.
- Hedging definition:
  - A client is considered to have a hedged FX position if its expected FX inflows, FX and FX-indexed claims cover at least 80 percent of its total expected FX outflows, FX and FX-indexed liabilities.
- Documentation:
  - A bank’s credit file must contain all relevant data for assessing identity, creditworthiness, regularity of settling liabilities and collateral quality, including approval and monitoring of retail credit exposures and analysis of borrower FX exposure.

### Benefits and limits of de-euroization under the peg
- Primary benefit:
  - Reducing balance sheet risks is the main advantage of de-euroization in Macedonia; it would allow exchange rate flexibility to serve as a “safety valve” in tail risk scenarios.
- Limited benefits under the peg:
  - Avoiding loss of seigniorage: data not available on seigniorage, and little evidence of significant currency substitution, so de-euroization is unlikely to affect seigniorage much.
  - Improving effectiveness of monetary policy: limited gains because Macedonia’s peg to the euro has served as the main policy anchor and the amount of monetary policy autonomy is limited.
- Macroeconomic context:
  - Given the peg, there may not be much gain in monetary policy transmission even with zero euroization.
  - From an individual banker’s perspective, perceived risk is small given the peg and potential government support in a tail risk depreciation.

### Cross-country lessons and policy implications for de-euroization
- Experiences from successful de-euroization cases:
  - Successful cases (Chile, Israel, Mexico, Poland) tended to have flexible exchange rate regimes.
  - De-euroization tends to be supported by: macroeconomic stability, active public debt management, and development of domestic financial markets.
  - Issuing local currency-denominated public debt can de-euroize the government’s balance sheet, lengthen the domestic currency yield curve and foster domestic debt markets.
  - De-euroization can be a long process and may continue after macro stability is achieved; indexation can persist (example: Chile).
- Prudential measures that can support de-euroization (examples from other countries):
  - Higher liquidity requirements on foreign currency deposits (Angola and Uruguay) or FX loans (Croatia).
  - Higher capital and provisioning requirements on foreign currency loans, especially to unhedged borrowers (Angola, Honduras, Vietnam).
  - Lower loan-to-value ratios on FX-denominated loans (Hungary).

### Conclusions and staff recommendations
- Persistence:
  - Even with incentives, Macedonia is likely to remain euroized for a long time; euroization is persistent regionally.
- Main recommendation:
  - Provide incentives and a supportive macro-financial context for gradual reduction of euroization through prudential policies, maintaining macro stability, and developing domestic debt markets.
- Specific policy actions urged by staff:
  - Maintain prudent macro policies and macro stability.
  - Improve public debt management and rely more on longer term denar loans to de-euroize the public sector balance sheet and lengthen the denar yield curve.
  - Develop a deep and liquid domestic capital market to provide investment opportunities in denar as substitutes for FX-denominated assets.

### Key numeric facts and thresholds cited
- 9.9 percent of total FX and FX-indexed loans (context: referenced in document header).
- Reserve requirements: 10 percent (denar deposits), 13 percent (FX deposits), 20 percent (FX-indexed deposits).
- Net open FX position limit: 30 per cent of the banks’ own funds.
- Capital charge multiplier for currency risk: 8%.
- Stress-test trigger for annual portfolio assessment: FX and FX-indexed loans to residents exceed 15 percent of total credit to residents (net of specified offsets).
- Hedged FX position definition threshold: 80 percent coverage of expected FX outflows by expected FX inflows and claims.
- De-euroization success criterion (from Reinhart, Rogoff, and Savastano (2003)): deposit euroization falls by at least 20 percentage points and remains below 20 percent.
- Peg history: pegged to the euro since January 2002; previously pegged to the Deutsch Mark since 1995.
- Historical stress periods summarized: drops in foreign assets averaged 17 percent and lasted between 6 and 14 months.

*IMF staff report excerpt (FYR Macedonia).*

### 7. The most severe crises in the last decade were aggravated by deposit

### _cr12134 - 7. The most severe crises in the last decade were aggravated by deposit

### Crises, deposits, and euroization
- Private sector deposits dropped by 14 percent in the security crisis of 2001 and by 8 percent in 2008–9.
- In both episodes the drop in corporate deposits was steeper and more persistent than household deposits.
- By the end of both crisis periods the degree of household euroization had increased significantly.
- Table excerpts (as reported):
  - Fe b 01 – Se p 01: 8 | 253.0 | 24.3 | 6.4 | 4.8
  - May 02 – Mar 03: 11 | 232.7 | 28.2 | 4.6 | 3.4
  - Oct 03 – Nov 04: 14 | 9.0 | 1.0 | 3.8 | 3.2
  - Nov 07 – Apr 08: 6 | 95.8 | 4.3 | 4.1 | 3.9
  - Oct 08 – May 09: 8 | 732.5 | 30.1 | 4.3 | 3.2
  - Note: Table header context: "Months of imports Cumulative loss in Reserves (mn dols) (percent*) at start at end" and source: NBRM. "* of FA on month prior to crisis"

### Monetary policy and defending the peg
- The NBRM consistently tightened the policy rate whenever reserves were “low and falling,” defending the peg.
- Specific policy actions and outcomes:
  - A large inflow of foreign reserves from the privatization of the telecommunications company in January 2001 kept foreign assets at comfortable levels even after a 24.3% loss in reserves during the security crisis in 2001; the NBRM reacted by more than doubling the policy rate to 20 percent.
  - During 2003–4 the drop in reserves was minor (1%) but import coverage collapsed to the historical minimum of 3.2 months; the NBRM raised the policy rate and a program with the IMF was agreed to strengthen the peg via increased foreign exchange reserves, relying on proceeds from two large privatizations in the first two years.
  - In the first half of 2008 the central bank tightened monetary policy despite relatively stable international reserves and broadly adequate import coverage; the explicit rationale was concerns that the economy was overheating with high inflation, high credit growth and current account deterioration.

### Complementary policy measures during stress episodes
- 2001:
  - Increased reserve requirements.
  - Imposed a temporary restriction on FX sales to big firms and bureaus.
  - Shortened the period for surrendering export proceeds from 180 to 30 days.
- 2003–4:
  - Introduced reserve requirements on FX deposits.
  - Shifted CB-bills auctions to volume tender.
- 2008:
  - Increased capital adequacy requirements for overdraft and credit card loans.
  - Broadened the base for reserve requirements.
- 2009:
  - Introduced liquidity ratios requirements.
  - Increased the reserve requirements for FX and FX indexed deposits.

### Degree of monetary autonomy and empirical proxies
- Imperfect capital mobility provides some short-term autonomy, but it appears limited.
- In a peg without autonomy UIP implies i = i* + premium; deviations from UIP indicate autonomy but the risk premium is not directly observable.
- Three proxies used for the risk premium term:
  - banks deposits and loans currency spreads,
  - household deposit euroization,
  - (implicit) currency spreads correlation with risk premium.
- Empirical observations:
  - The policy rate spread has historically tracked closely measures of risk premium.
  - The policy rate spread over Euro area rates has historically tracked measures of risk premium.
  - Recommendation: NBRM should continue timely monetary responses based on permanent monitoring of reserve adequacy measures, balance of payments dynamics, credit growth rates, trends in the ratio of euro-denominated to denar deposits, and Eurozone interest rates.

### Model of semi-autonomous monetary policy — structure and stylized results
- Model setup:
  - Central bank faces dual objective: defend the peg and affect inflation and output gap when the exchange rate is not under pressure.
  - Monetary autonomy assumed: domestic interest rates can differ from UIP.
  - Inflation (π_t) and output gap (x_t) follow new-Keynesian dynamics (Phillips curve and IS/aggregate demand).
  - In absence of exchange rate pressure the central bank follows a Taylor rule: i_t^* depends on (π_t − π^T) and x_t.
  - Reserve dynamics: Δr_t driven by deviations from UIP; reserves r_t must be kept above a threshold r̄.
  - Central bank minimizes a loss combining deviations of policy rate from Taylor rule and a reserve-related cost: min Σ (i_t − i_t^*)^2 + large cost as r approaches r̄, with reserve cost possibly asymmetric and depending on level and rate of change of reserves.
- Simplified, analytically tractable example assumptions:
  - Cut feedback from interest rates to inflation and output gap (treat inflation and output gap as exogenous).
  - Myopic central bank cares only about current constraint binding.
  - Reduced-form cost function with a weight θ_t on the Taylor-rule component that depends on reserve gap.
- Key calibration and implications:
  - The weight on the Taylor rule component becomes zero if reserves are 30 percent below the reserve target (calibration based on empirical NBRM reactions to ~30 percent reserve drops).
  - When reserves are below threshold (θ<1) the policy rate response to inflation or output shocks is muted (capping from below and/or above):
    - If reserve gap is negative (θ<1) and reserves are falling, a negative output gap or inflation shock leads to a smaller interest rate cut than under the Taylor rule (floor on policy rate, “capping from below”).
    - If reserves are accumulating, response to a positive inflation shock is more muted than under Taylor rule (ceiling on policy rate, “capping from above”).
  - Result: lower volatility of interest rates compared to the unconstrained (pure Taylor rule) case.
  - Note: the simplified example is myopic; a forward-looking central bank would internalize future constraint probabilities and act earlier to avoid hitting the lower bound.

### Conclusions and policy implications
- Despite an exchange rate peg, imperfect capital mobility allows some short-run monetary autonomy, but the latitude is limited.
- The policy rate spread over Euro area rates has closely tracked risk-premium measures, indicating constrained autonomy.
- Stylized model outcome: when a pegged central bank internalizes reserve constraints, interest rate volatility is lower than under a floating-rate Taylor-rule regime facing similar shocks.
- Main policy conclusions:
  - It is important that NBRM interest rate policies continue to be driven foremost by the need to preserve confidence in the peg by holding adequate international reserves.
  - Policy rate volatility in response to supply and demand shocks is limited in a country with an exchange rate peg compared to a country with similar shocks but a floating exchange rate.

*Source: NBRM data and analysis as presented in the provided content.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2012/_cr12134.pdf_