## _wp0977

## Source details

**Canonical URL:** [_wp0977](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0977.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0977.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0977.pdf.json)

---

### I. Introduction
- Many emerging market countries have increased holdings of international reserves; when insurance against sudden stops in external capital flows is introduced as a motive, these increases do not appear inordinately high (Jeanne, 2007).
- International reserves in the Eastern Caribbean Currency Union (ECCU) have increased; questions arise whether such levels are excessive given the opportunity cost of holding them.
- Comparison with similar economies shows reserves in the ECCU are not out of line; international reserves in the ECCU appear to be lower than in comparator countries.
- The ECCU has fully liberalized its capital account since 2005; insurance against adverse capital movements is an additional motive for holding reserves.
- Application of a variant of the Jeanne (2007) model to the ECCU yields:
  - Current reserve levels in the ECCU are adequate to cover a variety of external sector shocks, including shocks to the current account, and small to moderate capital outflows.
  - Moderate to severe capital outflows could be a challenge; severe capital outflows are probably once-in-a-century events and may be too costly to self-insure with greatly increased reserves.
  - Historically, reserve levels in the ECCU have been comfortably above the calculated optimal level, but over time actual reserves have converged on the optimal.
  - If the Eastern Caribbean Central Bank (ECCB) maintains its existing reserve cover rule (focused on demand liabilities), actual reserves could become inadequate (less than optimal).

### II. Traditional Measures of Reserve Adequacy
- Key functions of international reserves in the ECCU:
  - Buffer absorbing balance of payments shocks where access to borrowing is limited or costly, and the exchange rate is not used to ensure adjustment.
  - Provide confidence in the monetary authorities' commitment to support the currency.
- Common heuristic reserve adequacy measures:
  - Reserves equivalent to at least three months of imports of goods and services.
  - Reserves relative to the external current account balance or variability in foreign exchange revenues.
  - Ratios of reserves to monetary aggregates (reserve money or broad money).
  - Ratio of short-term external debt to reserves (benchmark value of one is a key vulnerability indicator).
- Empirical assessment for the ECCU at end-2007:
  - Foreign reserves in terms of months of imports were lower in the ECCU compared with other small tourism dependent economies, Caribbean neighbors, or monetary unions in Africa.
  - The ECCU has the lowest level of reserves in months of imports and relative to both broad money and reserve money.
- Note on models:
  - Buffer stock models (Frankel and Jovanovic, 1981) can explain ECCU reserve evolution (Grenade, 2008), but such models do not explicitly include the cost of external disequilibria or clearly define the opportunity cost of holding reserves.

### III. Optimal International Reserves (Jeanne (2007) insurance model)
- Model setup and objective:
  - Small open economy with representative consumer; wealth held as liquid international reserves (R) or higher-yielding illiquid asset (I).
  - A crisis is a loss of access to external credit that results in a fall in output.
  - Reserves smooth consumption during a crisis (crisis mitigation) and can reduce crisis probability (crisis prevention).
  - Holding reserves entails an opportunity cost ∂.
  - Optimal reserves minimize the consumer’s welfare loss function:
    - Loss = ∂R + π(R) f(R)
- Calibration and parameter choices for ECCU application:
  - Size of output loss ∆Y set to 10 percent of GDP.
  - Probability of a “sudden stop” estimated as a function of fundamentals via a probit (Jeanne, 2007); baseline assumes probability of a crisis is exogenous.
  - Opportunity cost of reserves set at ∂ = 3 percent.
  - Benchmark risk-aversion parameter set at σ = 2.
  - Data sources: international reserves, broad money, demand liabilities and output from the ECCB.
- Closed-form result (under assumption π independent of R):
  - Optimal reserves = “full insurance” level R = L + ∆Y minus a term reflecting the opportunity cost of holding reserves (Equation (5) referenced in source).

### IV. Current Account Shocks: ECCU vulnerabilities and model results
- ECCU exposures making current account shocks important:
  - Tourism receipts account for over 70 percent of total exports in the ECCU.
  - All oil and oil derivatives are imported; a large share of food consumption goods are imported.
  - High disaster proneness: between 1970 and 2002 ECCU countries were affected by 12 major natural disasters (mostly hurricanes), with a median impact on the current account of 10.8 percent of GDP.
- Model findings:
  - ECCU international reserves are close to or above the optimal level needed to insure the region against potential current account shocks.
  - Actual international reserves exceed the amount required to withstand a shock equivalent to:
    - three months of imports; and
    - a natural disaster of the average magnitude observed in the ECCU.
  - Example calibration: for a natural disaster with impact on the current account of 10.8 percent of GDP and a probability of occurring of 10 percent, the optimal level of reserves is around (figure referenced; numerical optimal not provided in supplied text).

### V. Capital account shocks: risks and reserve adequacy
- Character of capital account crises:
  - Sudden cessation or reversal of capital inflows forces large current account adjustments and large depreciations, associated with severe output contractions.
  - In the emerging market crises of the 1990s, the decline in real GDP growth averaged 10 percentage points in the most directly-affected countries; capital outflows in some cases amounted to as much as 15–20 percent of GDP (at annualized rates).
- ECCU structural changes affecting vulnerability:
  - Repeal of the Exchange Control Act in late 2005.
  - Foreign currency deposits in the banking sector expanded to about 17 percent of total deposits.
  - Public debt is largely externally funded and high relative to peers.
- Reserve benchmarks and interpretations:
  - If reserves equal or exceed short-term external debt, a country can meet external debt-servicing needs even without market access.
  - Augmenting reserves coverage with defense against sudden changes in foreign currency deposits provides additional protection where banking system dollarization is growing.
- Empirical assessment:
  - ECCB reserves are currently around the optimal size to insure the region against capital flight of the magnitude of all short-term external debt plus foreign currency deposits.
  - Actual reserves also cover a shock equivalent to all demand liabilities as expected under a currency board arrangement.
- Historical/numeric note preserved verbatim:
  - 8.5 percent of GDP. The actual level of reserves has consistently exceeded this level over the
- Figures referenced (percent of GDP): comparisons of actual reserves, 3 months of imports, 5 months of imports, demand liabilities, and foreign currency deposits plus short-term external debt illustrate reserve coverage over 1996–2008.

### VI. Lender-of-last-resort (LOLR) constraints and bank-run risk
- Institutional and operational constraints on ECCB acting as LOLR:
  - Quasi-currency board architecture requires monetary base to be covered by international reserves.
  - ECCB’s liquidity provision is bounded by international capital market access and tradeoffs between using international reserves and protecting the fixed exchange rate.
  - The Monetary Council can block LOLR use if lending may endanger exchange rate stability.
- Benchmark from Uruguay 2002:
  - After a severe external shock (spillover from Argentine crisis, output loss about 18 percent of GDP), about 40 percent of the banking system’s total deposits were withdrawn in 2002.
- Reserve sufficiency against deposit outflows:
  - Current international reserves are adequate to withstand an outflow equivalent to 20 percent of deposits.
  - Reserves would be insufficient in the event of a systemic shock that leads to withdrawal of 40 percent of banking system deposits.
  - Presence of strong foreign commercial banks could reduce reserve needs but requires coordinated contingency plans.

### VII. Jeanne (2007) insurance model: sensitivity analysis and scenarios
- Baseline calibration parameters and ranges:
  - Output loss ∆Y = 0.1; Range of variation [0,0.3]
  - Opportunity cost of reserves ∂ = 0.03; Range of variation [0.01,0.06]
  - Probability of a sudden stop π = 0.1; Range of variation [0,0.25]
  - Risk aversion σ = 2; Range of variation [1,10]
- Simulation outcomes:
  - To smooth the impact of a sudden stop equivalent to roll-over of ECCU external public debt plus outflow of all foreign currency deposits:
    - International reserves would have to increase to about 25 percent of GDP to cover an output loss of about 18 percent of ECCU GDP (Uruguay 2002-sized shock).
    - International reserves need to be about 15 percent of GDP if output loss is about 7½ percent of ECCU GDP.
- Cost of holding reserves:
  - Baseline opportunity cost amounts to 3 percent in yield forgone.
  - If cost of holding reserves increases to 4.5 percent, optimal reserves decline to about 12 percent of GDP compared with 17 percent of GDP in the baseline for 2008.
  - Optimal reserve holdings would reach 27 percent of GDP at end-2008 if holding cost declines to 0.5 percent of GDP.
- Alternative assessment of opportunity cost via marginal product of capital:
  - Preliminary estimate of the marginal product of capital in the ECCU is about 8 percent.
  - By this measure, holding reserves appears to carry an average net cost of about 1 percent of GDP (after netting out a short-term real return of about 2 percent in U.S. assets).

### VIII. Central bank reserve rule, money growth, and projections
- ECCB legal/operational reserve requirements:
  - By law, ECCB required to maintain international reserves equivalent to 60 percent of the central bank’s demand liabilities.
  - Operationally committed to maintaining a reserve cover greater than or equal to 80 percent of demand liabilities; in practice reserves have been close to 100 percent of demand liabilities, occasionally exceeding it.
- Financial deepening and divergence:
  - Broad money to GDP has risen since 1984, diverging from demand liabilities growth.
  - Estimated long-run income elasticities (Table 2):
    - Broad money income elasticity = 1.47 (*** significant at 1 percent level)
    - Demand liabilities income elasticity = 1.07 (*** significant at 1 percent level)
  - Long-run interest rate differential coefficients: -0.08 for broad money (***), -0.09 for demand liabilities (***).
  - Short-run dynamics: money returns to long-run equilibrium after about 10 quarters.
- Projections to 2015:
  - Projected reserves based on the 80 percent reserve cover rule continue to fall short of optimal reserves needed to withstand moderate-to-severe deposit withdrawals.
  - Even with 100 percent coverage of demand liabilities, the gap between reserves and potential bank deposit runs widens through the projection horizon.

### IX. Policy implications and recommendations
- Main findings:
  - Current ECCB reserve levels appear adequate for a variety of current account and capital account shocks using traditional ratios and optimal reserve frameworks.
  - Moderate-to-severe deposit withdrawals in the context of an open capital account could challenge the ECCB.
  - The ECCB’s 80 percent of demand liabilities rule has historically covered most shocks but cushion has decreased recently.
  - A reserve rule focused only on demand liabilities tends to under-insure against potential deposit outflows given financial deepening and capital account liberalization.
- Policy responses suggested:
  - Consider modifying the reserve cover rule to take into account the effects of capital account liberalization and some level of adverse capital movements.
  - Develop alternative shock absorbers to reduce reliance on reserves (especially if opportunity cost of holding reserves is high).
  - Strengthen contingency planning and regulatory framework for the banking system.
  - Establish credit lines with international financial institutions and other central banks to insure against large adverse capital movements.
  - Use interest rate on ECCB fixed deposits cautiously; raising it to attract commercial bank deposits could imply quasi-fiscal losses if ECCB cannot earn comparable returns on foreign reserves.
  - Retaining more of the central bank’s profits could help build reserves, but accumulation likely to be slow.

*Source: IMF working paper chapter “Comparison of International Reserves Across Regions”.*

### 1.         Comparison         of         International Reserves Across Regions..........................................

### Comparison of International Reserves Across Regions

### I. Introduction
- Many emerging market countries have increased holdings of international reserves; when insurance against sudden stops in external capital flows is introduced as a motive, these increases do not appear inordinately high (Jeanne, 2007).
- International reserves in the Eastern Caribbean Currency Union (ECCU) have increased; questions arise whether such levels are excessive given the opportunity cost of holding them.
- Comparison with similar economies (currency unions among developing countries, and tourism dependent economies) shows reserves in the ECCU are not out of line; in fact, international reserves in the ECCU appear to be lower than in comparator countries.
- The ECCU has fully liberalized its capital account since 2005; insurance against adverse capital movements is an additional motive for holding reserves.
- The paper applies a variant of the Jeanne (2007) model to the ECCU and finds:
  - Current reserve levels in the ECCU are adequate to cover a variety of external sector shocks, including shocks to the current account, and small to moderate capital outflows.
  - Moderate to severe capital outflows could be a challenge; severe capital outflows are probably once-in-a-century events and may be too costly to self-insure with greatly increased reserves.
  - Historically, reserve levels in the ECCU have been comfortably above the calculated optimal level, but over time actual reserves have converged on the optimal.
  - If the Eastern Caribbean Central Bank (ECCB) maintains its existing reserve cover rule (focused on demand liabilities), actual reserves could become inadequate (less than optimal).

### II. Traditional Measures of Reserve Adequacy
- Key functions of international reserves in the ECCU:
  - Buffer absorbing balance of payments shocks where access to borrowing is limited or costly, and the exchange rate is not used to ensure adjustment.
  - Provide confidence in the monetary authorities' commitment to support the currency.
- Common traditional/heuristic reserve adequacy measures:
  - Reserves equivalent to at least three months of imports of goods and services.
  - Reserves relative to the external current account balance or variability in foreign exchange revenues.
  - Ratios of reserves to monetary aggregates (reserve money or broad money) to indicate backing of central bank liabilities.
  - Ratio of short-term external debt to reserves (benchmark value of one is a key vulnerability indicator).
- Empirical assessment for the ECCU at end-2007:
  - Foreign reserves in terms of months of imports were lower in the ECCU compared with other small tourism dependent economies, Caribbean neighbors, or monetary unions in Africa.
  - The ECCU has the lowest level of reserves in months of imports and relative to both broad money and reserve money.
- Buffer stock models (Frankel and Jovanovic, 1981) can explain ECCU reserve evolution (Grenade, 2008), but such models do not explicitly include the cost of external disequilibria or clearly define the opportunity cost of holding reserves.

### III. Optimal International Reserves (Jeanne (2007) insurance model)
- Model setup and objective:
  - Small open economy with a representative consumer; wealth held as liquid international reserves (R) or higher-yielding illiquid asset (I).
  - A crisis is a loss of access to external credit that results in a fall in output.
  - Reserves smooth consumption during a crisis (crisis mitigation) and can reduce crisis probability (crisis prevention).
  - Holding reserves entails an opportunity cost ∂.
  - Optimal reserves minimize the consumer’s welfare loss function:
    - Loss = ∂R + π(R) f(R)
- Calibration and parameter choices used for ECCU application:
  - Size of output loss ∆Y set to 10 percent of GDP (in line with the average output cost of simultaneous currency and banking crises reported in the literature).
  - Probability of a “sudden stop” estimated as a function of fundamentals via a probit (Jeanne, 2007); baseline assumes probability of a crisis is exogenous.
  - Opportunity cost of reserves set at ∂=3 percent (historical average cost differential between issuing a ten-year U.S. treasury bond and investing in three-month U.S. Treasury bills).
  - Benchmark risk-aversion parameter set at σ=2.
  - Data sources for ECCU analysis: international reserves, broad money, demand liabilities and output from the ECCB.
- Closed-form result (under assumption π independent of R):
  - Equation (5) as presented:
    1
    11RL  Y
    σ
    π
    −
    ⎡⎤
    ∂
    ⎛⎞
    ⎢⎥
    =
    +Δ−−+
    ⎜⎟
    ⎢⎥
    ⎝⎠
    ⎣⎦
  - Interpretation: optimal reserves = “full insurance” level R = L + ∆Y minus a term reflecting the opportunity cost of holding reserves.

### IV. Current Account Shocks (ECCU vulnerabilities and model results)
- ECCU exposures that make current account shocks important:
  - Reliance on tourism: tourism receipts account for over 70 percent of total exports in the ECCU.
  - All oil and oil derivatives are imported; a large share of food consumption goods are imported.
  - High disaster proneness: between 1970 and 2002 ECCU countries were affected by 12 major natural disasters (mostly hurricanes), with a median impact on the current account of 10.8 percent of GDP.
- Model findings (summary of results reported):
  - ECCU international reserves are close to or above the optimal level needed to insure the region against potential current account shocks.
  - Actual international reserves exceed the amount required to withstand a shock equivalent to:
    - three months of imports; and
    - a natural disaster of the average magnitude observed in the ECCU.
  - Example calibration: for a natural disaster with impact on the current account of 10.8 percent of GDP and a probability of occurring of 10 percent, the optimal level of reserves is around (figure referenced but numerical optimal level not provided in the supplied text).

### V. Policy-relevant findings and implications
- Reserve adequacy conclusion:
  - Under baseline calibrations (∆Y = 10 percent of GDP, ∂=3 percent, σ=2), current ECCU reserves are generally adequate to self-insure against typical current account shocks and small-to-moderate capital outflows.
  - Moderate-to-severe capital outflows pose a significant challenge; severe outflows may be too costly to self-insure via much larger reserves.
- Institutional and rule-based considerations:
  - The ECCB is required to maintain international reserves to cover 60 percent of its demand liabilities and is often regarded as a quasi-currency board.
  - The ECCB’s reserve cover rule focuses on demand liabilities (a narrower operational concept) rather than the broader reserve requirements implied by insurance against capital flows.
  - Because the ECCB’s operational rule is narrower, actual reserves have historically converged toward the optimal; continued use of the same rule could lead to actual reserves becoming inadequate relative to optimal self-insurance needs.
- Sensitivity analysis (noted as explored in the paper):
  - The model’s sensitivity to the estimate of output loss ∆Y and the cost of holding reserves ∂ is examined in subsequent sections (Section IV).
  - Alternative measures for the opportunity cost of holding reserves include the marginal product of capital or the interest rate on external debt (Rodrik, 2006).

*Source: IMF working paper chapter “Comparison of International Reserves Across Regions”.*

### 8.5 percent of GDP. The actual level of reserves has consistently exceeded this level over the

### _wp0977 - 8.5 percent of GDP. The actual level of reserves has consistently exceeded this level over the

### Capital account shocks: risks and reserve adequacy
- Capital account crises feature sudden cessation or reversal of capital inflows, forcing large current account adjustments and large depreciations, and have been associated with severe output contractions.
- In the emerging market crises of the 1990s, the decline in real GDP growth averaged 10 percentage points in the most directly-affected countries; capital outflows in some cases amounted to as much as 15–20 percent of GDP (at annualized rates).
- Short-term adjustment channels: import compression, slump in domestic demand, and balance sheet effects from exchange rate and interest rate exposures.
- Recent structural changes in the ECCU:
  - Repeal of the Exchange Control Act in late 2005.
  - Foreign currency deposits in the banking sector expanded to about 17 percent of total deposits.
  - Public debt is largely externally funded and high relative to peers.
- Reserve benchmarks and interpretations:
  - If reserves equal or exceed short-term external debt, a country can meet external debt-servicing needs even without market access.
  - Augmenting reserves coverage with defense against sudden changes in foreign currency deposits provides additional protection where banking system dollarization is growing.
- Empirical assessment:
  - ECCB reserves are currently around the optimal size to insure the region against capital flight of the magnitude of all short-term external debt plus foreign currency deposits.
  - Actual reserves also cover a shock equivalent to all demand liabilities as expected under a currency board arrangement.
- Figures referenced (in percent of GDP): comparisons of actual reserves, 3 months of imports, 5 months of imports, demand liabilities, and foreign currency deposits plus short-term external debt illustrate reserve coverage over 1996–2008.

### Lender-of-last-resort (LOLR) constraints and bank-run risk
- Institutional and operational constraints on ECCB acting as LOLR:
  - Quasi-currency board architecture requires monetary base to be covered by international reserves.
  - ECCB’s liquidity provision is bounded by international capital market access and tradeoffs between using international reserves and protecting the fixed exchange rate.
  - Monetary Council (Ministers of Finance of member countries) can block LOLR use if lending may endanger exchange rate stability.
- Benchmark from Uruguay 2002: after a severe external shock (spillover from Argentine crisis, output loss about 18 percent of GDP), about 40 percent of the banking system’s total deposits were withdrawn in 2002.
- Reserve sufficiency against deposit outflows:
  - Current international reserves are adequate to withstand an outflow equivalent to 20 percent of deposits.
  - Reserves would be insufficient in the event of a systemic shock that leads to withdrawal of 40 percent of banking system deposits.
  - Presence of strong foreign commercial banks could reduce reserve needs but requires coordinated contingency plans.

### Jeanne (2007) insurance model: sensitivity analysis and scenarios
- Baseline calibration parameters (from Jeanne (2007)) shown in source:
  - Output loss ∆Y = 0.1; Range of variation [0,0.3]
  - Opportunity cost of reserves ∂ = 0.03; Range of variation [0.01,0.06]
  - Probability of a sudden stop π = 0.1; Range of variation [0,0.25]
  - Risk aversion σ = 2; Range of variation [1,10]
- Simulation outcomes:
  - To smooth the impact of a sudden stop equivalent to roll-over of ECCU external public debt plus outflow of all foreign currency deposits:
    - International reserves would have to increase to about 25 percent of GDP to cover an output loss of about 18 percent of ECCU GDP (Uruguay 2002-sized shock).
    - International reserves need to be about 15 percent of GDP if output loss is about 7½ percent of ECCU GDP.
- Cost of holding reserves:
  - Baseline opportunity cost amounts to 3 percent in yield forgone.
  - If cost of holding reserves increases to 4.5 percent, optimal reserves decline to about 12 percent of GDP compared with 17 percent of GDP in the baseline for 2008.
  - Optimal reserve holdings would reach 27 percent of GDP at end-2008 if holding cost declines to 0.5 percent of GDP.
- Alternative assessment of opportunity cost via marginal product of capital:
  - Preliminary estimate of the marginal product of capital in the ECCU is about 8 percent.
  - By this measure, holding reserves appears to carry an average net cost of about 1 percent of GDP (after netting out a short-term real return of about 2 percent in U.S. assets).

### Central bank reserve rule, money growth, and projections
- ECCB legal/operational reserve requirements:
  - By law, ECCB required to maintain international reserves equivalent to 60 percent of the central bank’s demand liabilities.
  - Operationally committed to maintaining a reserve cover greater than or equal to 80 percent of demand liabilities; in practice reserves have been close to 100 percent of demand liabilities, occasionally exceeding it.
- Financial deepening and divergence:
  - Rapid financial deepening: broad money to GDP has risen since 1984, diverging from demand liabilities growth.
  - Estimated long-run income elasticities (Table 2):
    - Broad money income elasticity = 1.47 (*** significant at 1 percent level)
    - Demand liabilities income elasticity = 1.07 (*** significant at 1 percent level)
  - Long-run interest rate differential coefficients: -0.08 for broad money (***), -0.09 for demand liabilities (***).
  - Short-run dynamics: money returns to long-run equilibrium after about 10 quarters.
- Projections to 2015 using estimated coefficients, medium-term framework for ECCU Common Policies, and WEO US interest rate projections:
  - Projected reserves based on the 80 percent reserve cover rule continue to fall short of optimal reserves needed to withstand moderate-to-severe deposit withdrawals.
  - Even with 100 percent coverage of demand liabilities, the gap between reserves and potential bank deposit runs widens through projection horizon.

### Policy implications and recommendations
- Main findings:
  - Current ECCB reserve levels appear adequate for a variety of current account and capital account shocks using traditional ratios and optimal reserve frameworks.
  - However, moderate-to-severe deposit withdrawals in the context of an open capital account could challenge the ECCB.
  - The ECCB’s 80 percent of demand liabilities rule has historically covered most shocks but cushion has decreased recently.
  - A reserve rule focused only on demand liabilities tends to under-insure against potential deposit outflows given financial deepening and capital account liberalization.
- Policy responses suggested in the text:
  - Consider modifying the reserve cover rule to take into account the effects of capital account liberalization and some level of adverse capital movements.
  - Develop alternative shock absorbers to reduce reliance on reserves (especially if opportunity cost of holding reserves is high).
  - Strengthen contingency planning and regulatory framework for the banking system.
  - Establish credit lines with international financial institutions and other central banks to insure against large adverse capital movements.
  - Use interest rate on ECCB fixed deposits cautiously; raising it to attract commercial bank deposits could imply quasi-fiscal losses if ECCB cannot earn comparable returns on foreign reserves.
  - Retaining more of the central bank’s profits could help build reserves, but accumulation likely to be slow.

*Source: IMF Fund staff calculations and Jeanne (2007) as presented in the supplied content.*

### References

### _wp0977 - References

### Reserve adequacy, international reserves, and optimal reserves
- Aizenman, J. and N. Marion, 2004, International Reserve Holdings with Sovereign Risk and Costly Tax Collection”, The Economic Journal, Vol. 114, pp. 569–91.
- Delèchat, C. and J.K. Martijn, 2007, “Reserve Adequacy in the CFA Franc Zone,” in A.M. Gulde and C. Tsangarides (editors), The CFA Franc Zone. Common Currency, Uncommon Challenges, Washington DC: International Monetary Fund.
- Drummond, P. and A. Dhasmana, 2008, “Foreign Reserve Adequacy in Sub-Saharan Africa,” IMF Working Paper 08/150, Washington DC: International Monetary Fund.
- Floerkemeir H. and M. Sumlinski, 2008, “International Reserve Trends in the South Caucasus and Central Asia Region,” IMF Working Paper 08/41, Washington DC: International Monetary Fund.
- Flood, R. and N. Marion, 2002, “Holding International Reserves in an Era of High Capital Mobility,” IMF Working Paper 02/62, Washington DC: International Monetary Fund.
- Frankel, J. and B. Jovanovic, 1981, “Optimal International Reserves: A Stochastic Framework,” The Economic Journal, Vol. 91, pp. 507–14.
- Gonçalves, F.M., 2007, “The Optimal Level of Reserves in Financially Dollarized Economies: The Case of Uruguay.” IMF Working Paper 07/625, Washington DC: International Monetary Fund.
- Goyal, R. and P. Liu, 2006, “Foreign Exchange Reserve Accumulation in Brazil”, mimeo, IMF Western Hemisphere Department. Washington DC: International Monetary Fund.
- Jeanne, O. and R. Rancière, 2006, “The Optimal Level of Reserves For Emerging Markets: Formulas and Applications,” IMF Working Paper 06/229, Washington DC: International Monetary Fund.
- Jeanne, O., 2007, “International Reserves In Emerging Market Countries: Too Much of a Good Thing?” Brookings Papers in Economic Activity, 1: 2007, pp. 1–79.
- Rodrik, D., 2006, “ The Social Cost of Foreign Exchange Reserves,” International Economic Journal, Vol. 15, pp. 253–66.
- Wijnholds, J., and A. Kapteyn, 2001, “Reserve Adequacy in Emerging Market Economies”, IMF Working Paper 01/143, Washington DC: International Monetary Fund.

### Lender of last resort, currency crises, and systemic banking crises
- Calvo, G., 2006, “Monetary Policy Challenges in Emerging Markets: Sudden Stop, Liability Dollarization, and Lender of last Resort,” Working Paper 12788. Cambridge, Mass.: National Bureau of Economic Reserch (December).
- Dehesa, M. and P. Druck, 2008, “The Eastern Caribbean Central Bank: Challenges to An Effective Lender of Last Resort,” IMF Working Paper 08/214, Washington DC: International Monetary Fund.
- Jeanne, O. and C. Wyplosz, 2003, “The International Lender of  Last Resort: How Large is Large Enough?” In Managing Currency Crises in Emerging Markets, edited by Michael P. Dooley and Jeffry A. Frankel, Chicago: University of Chicago Press.
- Hoelscher, D.S. and M. Quityn, 2003, Managing Systemic Banking Crises, Occasional Paper 224, Washington DC: International Monetary Fund.
- Sachs, J., A. Tornell, and Z.A. Velasco, 1996, “Financial Crisis in Emerging Markets: The Lessons from 1995,” Brookings Papers on Economic Activity, Vol.1, pp. 177–215.

### Caribbean, Eastern Caribbean, and regional integration
- Drummond, P. and A. Dhasmana, 2008, “Foreign Reserve Adequacy in Sub-Saharan Africa,” IMF Working Paper 08/150, Washington DC: International Monetary Fund. 
- Grenade, K., 2008, “The Optimal Level of Foreign Reserves for the ECCB,” mimeo, Research Department, Eastern Caribbean Central Bank, St. Kitts and Nevis.
- Kwon, G., S. Panth, J. Chai, A. Gerson, G.M. Casselle, R. Espinoza, 2008, “Financial Integration in the Caribbean,” in A. Bauer, P. Cashin and S. Panth (editors), The Caribbean: Enhancing Economic Integration, IMF Western Hemisphere Department, Washington DC: International Monetary Fund, pp. 3-43.
- Randall, R., 2006, “Eastern Caribbean Tourism: Developments and Outlook,” in R. Sahay, D.O. Robinson and P. Cashin (editors), The Caribbean: From Vulnerability to Sustained Growth, Washington DC: International Monetary Fund, pp. 285–306.
- Rasmussen, T., 2006, “Natural Disasters and their Macroeconomic Implications,” in R. Sahay, D.O. Robinson and P. Cashin (editors), The Caribbean: From Vulnerability to Sustained Growth, Washington DC: International Monetary Fund, pp. 181–205.
- Sahay, R., 2006, “Stabilization, Debt, and Fiscal Policy in the Caribbean,” in R. Sahay, D.O. Robinson and P. Cashin (editors), The Caribbean: From Vulnerability to Sustained Growth, Washington DC: International Monetary Fund, pp. 17–57.
- Dehesa, M. and P. Druck, 2008, “The Eastern Caribbean Central Bank: Challenges to An Effective Lender of Last Resort,” IMF Working Paper 08/214, Washington DC: International Monetary Fund.
- Goyal, R. and P. Liu, 2006, “Foreign Exchange Reserve Accumulation in Brazil”, mimeo, IMF Western Hemisphere Department. Washington DC: International Monetary Fund.

### Money demand, term premia, and macroeconomic methods
- Judd, J.P. and J.L. Scadding, 1982, “The Search for a Stable Money Demand Function: A Survey of the Post 1973 Literature,” Journal of Economic Literature, Vol. XX, pp. 993–1023.
- Taufiq, C., 2002, “Financial Innovations and the Demand for United States M1 and M2 Components”, International Economic Journal, March, 2002.
- Rudebusch, G., B.P. Sack, and E.T. Swanson, 2007, “Macroeconomic Implications of Changes in the Term Premium,” Federal Reserve Bank of St. Louis Review Vol. 89, No. 4, pp. 241–69.
- Lucas, R.E., 1987, Models of Business Cycles, Oxford: Blackwell.
- Pemberton, J., 1996, “Growth Trends, Cyclical Fluctuations, and Welfare With Non-expected Utility Preferences” Economic Letters Vol. 50, pp.387–92.

### Crisis costs, output costs, and social cost analysis
- Hutchison, M. and I. Noy, 2005, “How Bad are the Twins? Output Costs of Currency and Banking Crisis,” Journal of Money,Credit and Banking, Vol. 37, no.4 (August).
- Sachs, J., A. Tornell, and Z.A. Velasco, 1996, “Financial Crisis in Emerging Markets: The Lessons from 1995,” Brookings Papers on Economic Activity, Vol.1, pp. 177–215.
- Rodrik, D., 2006, “ The Social Cost of Foreign Exchange Reserves,” International Economic Journal, Vol. 15, pp. 253–66.

*Source: _wp0977 - References*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0977.pdf_