## wpiea2019058

## Source details

**Canonical URL:** [wpiea2019058](https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019058.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2019/wpiea2019058.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2019/wpiea2019058.pdf.json)

---

### Major findings on fixed exchange rate arrangements and scope
- "76 countries’ exchange rate arrangements out of the then-192 IMF member countries belonged to the category of fixed exchange rates," including currency boards, conventional pegs, stabilized arrangements, and crawling pegs (AREAER 2016).
- "71 IMF members operated market determined exchange rate arrangements, such as floating and free floating."
- Core focus: currency boards and pegs to a single currency (conventional pegs), with broader relevance to fixed exchange rates.
- Fixed regimes are heterogeneous: include small open economies, advanced economies, emerging markets, low-income countries; commodity exporters with low economic diversification form a significant share.
- Commitment heterogeneity: currency boards represent strongest commitment; some pegged regimes regularly resort to devaluations.

### Theoretical underpinnings and conceptual framework
- Optimal regime criteria emphasize: small economy size, high trade integration, flexible labor markets, sound fiscal policy, shocks that are nominal or symmetric with anchor country.
- Following Mundell (1963), optimal exchange rate choice depends on nature of shocks (real vs nominal) and capital mobility.
- Under a peg with open capital account, pegging to a low-inflation currency can serve as a credible nominal anchor where domestic monetary credibility is weak.
- Price-specie flow mechanism extended to open-economy balance of payments with financial and capital accounts:
  - FX surpluses → banks sell FX to central bank (NFA ↑, bank reserves ↑) or capital outflows.
  - FX shortages → banks purchase FX (NFA ↓, bank reserves ↓) or capital inflows.
  - Bank reserves ↔ domestic money market rates: reserves ↓ → rates ↑ (may trigger capital inflows); reserves ↑ → rates ↓ (may trigger capital outflows).
- Paper models liquidity dynamics via bank reserves and short-term interest rates rather than goods-market price adjustments.

### Reserve demand and supply characterization
- Reserve demand:
  - Inversely related to short-term interest rates; precautionary reserves persist even without formal reserve requirements.
  - Minimum precautionary reserve level depends on money market liquidity, depth, efficiency, and payment system.
- Reserve supply:
  - Without central bank intervention, reserve supply is driven by autonomous factors—especially central bank NFA—and can be endogenous under a peg with open capital account.
  - Where government account at central bank is low/stable and CiC negligible, NFA largely determine bank reserve supply absent OMOs.
  - Change in NFA reflects current account and capital account balances; capital flows depend on interest rate differential with anchor currency.
- Assumed reserve supply curve: upward sloping—"an increase in the domestic money market rate attracts more FX and increases the NFA and bank reserves"—slope driven by capital flow elasticity to the domestic rate.
  - Capital flow elasticity influenced by country-specific "pull factors" and external "push factors" such as global liquidity and risk aversion, and can change over time.

### Money market equilibrium mechanics and key relationships
- No-central-bank-operations benchmark (currency board rule, full capital mobility):
  - FXI = R - R1
  - FXO = R2 - R
  - i = i* + Prem
  - With frictions, FX flows respond only when domestic rates deviate by more than +/- f from i*.
- Adjustment examples:
  - Reserve shortage R1<R → domestic short-term rates ↑ → FX inflows FXI = R - R1 offset shortage.
  - Reserve surplus R2>R → domestic rates ↓ → FX outflows FXO = R2 - R offset surplus.
- Vulnerabilities:
  - Friction costs, market segmentation, settlement delays, time-zone differences, country/counterparty risk impair seamless adjustment.
  - Volatile short-term rates ↑ precautionary reserve demand, impair money market development.
  - NFA can swing widely from ToT shocks or sudden capital outflows; persistent liquidity shortfalls can force high domestic rates and threaten banking solvency; persistent excess reserves deter market development.

### Operational implications and liquidity management objectives
- Monetary policy implementation goals under a peg with open capital account:
  - Defend the peg.
  - Preserve financial stability.
  - Steer short-term money market rates to stabilize capital flows and reduce FX intervention.
  - Reduce impact of autonomous factor fluctuations on bank reserves to limit interest rate volatility.
- Operational implications:
  - OMOs should be calibrated on forecasts of autonomous factors to stabilize reserves and interest rates.
  - Provide offsetting operations when autonomous factors or exogenous shocks (e.g., ToT shocks) would produce destabilizing reserve and rate fluctuations.
  - Recognize strong interaction between policy rate changes, capital flows, and reserves under a peg.

### Tests and operational equations used for offset assessment
- Central bank balance sheet identity disaggregated:
  - NFA = R + LA – LP + CiC + GOV + OIN
    - R = bank reserves at central bank
    - LA = liquidity absorption operations
    - LP = liquidity-providing operations
    - CiC = currency in circulation
    - GOV = government account at central bank
    - OIN = other items
- Estimated first-difference equations used:
  - Equation (2): ∆NFA_i,t = α1 ∆i_i,t + α2 ∆R_i,t + α3 ∆LP_i,t + α4 ∆LA_i,t + α5 ∆CiC_i,t + α6 ∆GOV_i,t + ε_t
    - Test 1: H0 α1 = 0 (central bank does not use policy rate to offset NFA changes).
  - Equation (3): ∆LP_i,t = α1 ∆i_i,t + α2 ∆R_i,t + α3 ∆NFA_i,t + α4 ∆LA_i,t + α5 ∆CiC_i,t + α6 ∆GOV_i,t + ε_t
    - Test 2: H0 α3 = 0 (central bank does not adjust liquidity-providing OMO to sterilize NFA changes).
  - Equation (4): ∆LA_i,t = α1 ∆i_i,t + α2 ∆R_i,t + α3 ∆LP_i,t + α4 ∆NFA_i,t + α5 ∆CiC_i,t + α6 ∆GOV_i,t + ε_t
    - Test 3: H0 α4 = 0 (central bank does not adjust liquidity-absorbing OMO to sterilize NFA changes).

### Instruments, operational choices, and trade-offs
- Main instruments to offset NFA fluctuations:
  - Policy rate: useful when capital account shocks are responsive to the policy rate.
  - Reserve requirements (with averaging): useful when capital flows unresponsive to policy rate and to absorb unexpected shocks.
  - OMOs (repo, bills, CDs) and standing facilities: used to inject/absorb liquidity; frequency and design matter (weekly OMOs often preferred).
- Design considerations:
  - Full-allotment auctions can defeat shock-offsetting objectives and risk over-injection.
  - Structural liquidity surplus calls for issuance of central bank securities to sterilize excess liquidity.
  - Reserve averaging periods of about one month reduce need for perfect forecasts; short averaging (≤ one week) reduces buffering capacity.
  - Standing facility spreads should exceed transaction costs to encourage interbank trading and limit SF recourse.
  - Collateral frameworks matter: large stock of government securities preferred; where shallow, broaden eligible collateral with haircuts and risk controls.

### Country practices and examples (selected)
- Denmark (Danmarks Nationalbank, DNB):
  - ERM central rate 746.038 krone per 100 euro, band +/- 2.25 percent.
  - Policy rates: lending rate; CDs rate; discount rate; current-account rate; OMOs weekly (usually seven-day CDs).
  - DNB fixes OMO rates; counterparties determine volumes.
  - Econometric findings (2000Q4–2017Q4):
    - Using interest rate differential (DNB lending rate minus ECB policy rate, i-i*): a 1 percentage point increase in the interest rate differential at t-3 increases ∆NFA by DKK 2.7 billion (statistically significant at 90 percent confidence, p-value < 0.1).
    - Substituting ECB policy rate: change in ECB policy rate at t-3 coefficient is -4.930738* (Table 2).
    - DNB sterilizes NFA effects on bank reserves by adjusting volumes of monetary operations (DLP decreases, CDs outstanding increase; statistically significant at p-value < 0.01).
    - Reserve supply curve observations: downward sloping during 2000–06 and 2007–14; upward sloping in 2015–17.
- Hong Kong SAR (HKMA):
  - Linked Exchange Rate System at 7.8 HKD per one USD since October 17, 1983.
  - Currency board implication: monetary base fully backed by foreign reserves; monetary base changes passively upon banks’ requests.
  - Base rate introduced September 1998; base rate currently set at 50 basis points above prevailing US Fed Fund target rate or the average of the five-day moving averages of the O/N and one-month HIBORs, whichever is higher.
  - Empirical findings (2001Q3–2017Q4):
    - Equation (2): change in HKMA base rate at t-1 coefficient 15597.83** (6532.251) associated with DNFA (R-squared 0.701477); HKMA notes significance does not imply it uses base rate to react to FX reserves.
    - Equation (3): discount window volume not statistically responsive to NFA.
    - Equation (4): increase in NFA results in increase in Exchange Fund bills and notes issued; HKMA states issuance meets banks’ demand for liquidity management and reflects composition shift of monetary base.
- Saudi Arabia (SAMA):
  - O/N reverse repo rate closely tracks Fed Fund target (lower bound).
  - Structural liquidity surplus due to oil-related flows; SAMA issues SAMA bills with one-, four-, 13-, 26-, and 52-week maturities to sterilize excess liquidity.
  - In 2016 liquidity tightening response: eased loan-to-deposit ratio; reduced weekly issuance of SAMA bills from 9 billion SAR to 3 billion SAR; offered term repurchase facilities and placements of deposits.
  - Variable width interest rate corridor: floor equal to Fed Funds Target (lower bound) since mid-2009; ceiling near 2 percent from 2009 until March 2018, then gradually increased to 3 percent currently.

### Empirical estimates—selected coefficients and statistics (preserved)
- Denmark—Table highlights (Appendix VI):
  - IDIFF(-3): 2.695068*** (1.502025) — a 1 percentage point increase in interest differential at t-3 → ∆NFA DKK 2.695068 billion (statistical annotation: *** p < 0.1).
  - DIF(-3) (ECB policy rate change): -4.930738* (1.578297).
  - DR coefficient in DNFA regressions: 1.034366* (0.023093).
  - DLP coefficient in DNFA regressions: -1.038834* (0.022616).
  - DLA coefficient in DNFA regressions: 0.965646* (0.019960).
  - DCIC coefficient in DNFA regressions: 0.934669* (0.282874).
  - DGOV coefficient in DNFA regressions: 0.970327* (0.017615).
  - R-squared 0.992165; Adjusted R-squared 0.991543 (Table 1).
  - Table 2 R-squared 0.992886; Adjusted R-squared 0.992312.
  - DLP regression R-squared 0.976974; Adjusted R-squared 0.975579.
  - DLA regression R-squared 0.985630; Adjusted R-squared 0.984759.
- Hong Kong SAR—Appendix VII:
  - DI(-1) (base rate change at t-1): 15597.83** (6532.251) in DNFA regression; R-squared 0.701477; Adjusted R-squared 0.687032.
  - DR coefficient in DLA regression: -0.331331* (0.079627).
  - DNFA coefficient in DLA regression: 0.418733* (0.043389).
  - DGOV coefficient in DLA regression: -0.159675** (0.064356).
  - DLA regression R-squared 0.499424; Adjusted R-squared 0.483781.

### Practical guidance and policy recommendations
- Central banks under fixed exchange rates and open capital accounts should aim to:
  - Use an adequate combination of instruments (reserve requirements with averaging, interest rate aligned to anchor adjusted by risk premium, OMOs calibrated on autonomous factors’ forecasts, and standing facilities) to offset short-term liquidity shocks.
  - Calibrate OMOs on autonomous factor forecasts to dampen interest rate volatility and foster money market development (weekly OMOs often preferred).
  - Use reserve averaging periods (around one month) to absorb unexpected shocks; remunerate required reserves close to opportunity cost when ratio is low and objective is not to curb credit growth.
  - Ensure standing facility spreads exceed transaction costs to promote interbank trading and limit facility recourse.
  - Maintain a collateral framework broad enough to support refinancing and emergency liquidity assistance when government securities markets are shallow.
  - Limit banks’ excess reserves to reduce the financing source for large FX purchases and speculation; where necessary, sterilize surpluses via issuance of central bank securities rather than relying solely on reserve requirements.
  - Set policy rate close to anchor country policy rate adjusted for country-specific risk premium when policy-rate-based offsetting is feasible; use indicators like O/N interest rate swaps and CDS spreads to assess anchor and risk premium.

### Consolidated conclusions
- Theoretical framework shows liquidity management under a peg can be automatic via FX flows, but frictions and autonomous factor shocks make central bank intervention justified to stabilize short-term rates and preserve financial stability.
- Empirical cross-country evidence:
  - Majority use reserve requirements (often ~one month averaging).
  - Many use short-term OMOs, but not always calibrated to autonomous factor forecasts.
  - Few maintain an interest rate corridor; in some cases policy rate is used to offset FX flows.
- With calibrated instruments and forecasting, central banks can offset short-term shocks, dampen interest rate volatility, lower precautionary reserve demand and funding costs, and build/release liquidity buffers to manage reversals.

*Source: IMF staff compilation of the content unit wpiea2019058 (text and figures as provided).*

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

### wpiea2019058 - References

### Major findings on fixed exchange rate arrangements and scope
- According to the Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER 2016), "76 countries’ exchange rate arrangements out of the then-192 IMF member countries belonged to the category of fixed exchange rates," including currency boards, conventional pegs, stabilized arrangements, and crawling pegs.
- "71 IMF members operated market determined exchange rate arrangements, such as floating and free floating."
- Core target group of the paper: currency boards and pegs to a single currency (conventional pegs), with relevance for the broader group of fixed exchange rates.
- Fixed exchange rate regimes are heterogeneous: they include small open economies as well as advanced economies, emerging markets, and low-income countries; commodity exporters with low economic diversification form a significant share.
- Commitment heterogeneity: currency boards represent the strongest commitment; some pegged regimes regularly resort to devaluations.

### Theoretical underpinnings and criteria for regime choice
- Academic criteria for choosing exchange rate arrangements emphasize: small economy size, high trade integration, flexible labor markets, sound fiscal policy, shocks that are nominal or symmetric with anchor country.
- Following Mundell (1963) and related literature, optimal exchange rate choice depends on the nature of shocks (real vs nominal) and capital mobility.
- Pegged regimes can be attractive as a credible nominal anchor when pegging to a low-inflation currency is a pre-commitment to anchor inflation expectations, particularly where independent monetary policy credibility is weak.

### Paper focus and objectives
- Focus: challenges of implementing monetary policy under a fixed exchange rate arrangement that does not rely on capital controls or devaluations.
- Objective: propose a liquidity management framework under normal conditions for central banks with fixed exchange rate arrangements and open capital accounts, operationalizing an interest rate corridor for fixed exchange rates without capital controls.
- Monetary policy implementation goals under a peg with open capital account:
  - Defend the peg.
  - Preserve financial stability.
  - Steer short-term money market rates to stabilize capital flows and reduce need for FX intervention.
  - Reduce the impact of autonomous factor fluctuations on bank reserves to limit interest rate volatility.

### Key challenges for liquidity management under a peg
- Forecasting autonomous factors in the face of sizeable current account and/or capital account shocks.
- Calibrating open market operations (OMOs) based on autonomous factor forecasts.
- Sterilizing large liquidity shocks induced by Terms-of-Trade (ToT) shocks or capital flows.
- Managing risks of currency speculation, especially around anticipated devaluation or misalignment.

### Conceptual framework highlights
- Revisit of the price-specie flow mechanism (Hume 1752) extended to an open-economy balance of payments including financial and capital accounts.
- Under a peg and open capital account:
  - FX surpluses lead to banks selling FX to the central bank (increasing NFA and bank reserves) or to capital outflows.
  - FX shortages lead to banks purchasing FX (decreasing NFA and bank reserves) or capital inflows.
  - Changes in bank reserves affect domestic money market rates; lower reserves raise rates (possibly triggering capital inflows), higher reserves lower rates (possibly triggering capital outflows).
- The paper models liquidity dynamics via fluctuations in bank reserves and effects on short-term interest rates rather than via goods market price adjustments.

### Reserve demand and supply characterization
- Reserve demand:
  - Follows standard intuition that demand for bank reserves is inversely related to short-term interest rates, reflecting precautionary reserve needs even with no reserve requirements.
  - Minimum precautionary reserves depend on money market liquidity, depth, efficiency, and the payment system.
- Reserve supply:
  - In the absence of central bank intervention, reserve supply becomes a function of autonomous factors—especially central bank NFA—and can be endogenous under a peg with open capital account.
  - Where the government account at the central bank is low/stable and currency in circulation (CiC) is negligible, NFA largely determine bank reserve supply absent central bank OMOs.
  - Change in NFA reflects current account and capital account balances; capital flows depend on the interest rate differential with the anchor currency.
- Theoretical assumption adopted: an upward sloping reserve supply curve—"an increase in the domestic money market rate attracts more FX and increases the NFA and bank reserves"—with the slope driven by capital flow elasticity to the domestic money market rate.
  - Capital flow elasticity is influenced by country-specific "pull factors" and external "push factors" such as global liquidity and risk aversion, and can change over time.
  - The paper later (Section IV) presents empirical illustration of shifting reserve supply curves using Denmark’s experience.

### Operational implications (conceptual)
- Monetary operations under a peg should aim to:
  - Steer short-term money market rates to a level that balances capital flow pressures and reduces FX intervention needs.
  - Use OMOs calibrated on forecasts of autonomous factors to stabilize reserves and interest rates.
  - Provide offsetting operations when autonomous factors or exogenous shocks (for example, ToT shocks) would otherwise produce destabilizing reserve and interest rate fluctuations.
- Recognition that modern monetary policy focuses on policy rate changes, but under a peg with open capital account such rate changes interact strongly with capital flows and reserves.

*Source: IMF working paper text (excerpt).*

### 1.1 Money Market Equilibrium under Fixed

### 1.1 Money Market Equilibrium under Fixed Exchange Rate with No Central Bank Monetary Operations

### C. The Assumption of Seamless Liquidity Adjustment by Capital Flows
- Framework: central bank operates under a currency board rule; local currency (CiC and bank reserves at the central bank) issued only against FX reserves. Government has an account at the central bank; central bank has no claims on the government. Banks not subject to reserve requirements. Banks rely on FX flows in liquidity management.
- Central bank balance sheet driven by FX reserves (main asset). Liabilities: CiC, government account, bank reserves, and other items (net).
- Two sets of factors impacting bank reserves (aggregated position R):
  - Autonomous factors: CiC, the government account, and the NFA.
  - Banks’ precautionary demand for reserves: function of settlement risk and the level of development and sophistication of the money markets.
- No-friction market assumptions initially:
  - Banks can buy and sell FX freely with no conversion costs; no country or counterparty risk; local banks can borrow and lend FX under the same conditions as foreign counterparties.
  - Under full capital mobility, reserve shortages or surpluses are offset by capital flows; short-term rates respond to reserve deviations and gravitate toward the interest rate parity.
- Adjustment mechanics (as presented):
  - Any decrease in bank reserves at the central bank to R1 below equilibrium (R1<R) will increase domestic short-term rates and trigger FX inflows (FXI) that offset the shortage: FXI = R - R1.
  - Any increase of bank reserves above equilibrium (R2>R) will decrease short-term rates and trigger FX outflows (FXO) that offset the surplus: FXO = R2 - R.
- Conditions required for the smooth adjustment mechanism:
  - Banks use FX inflows/outflows for liquidity adjustment.
  - FX flows respond efficiently to changes in domestic money market rates (no settlement delays or time-zone frictions; short-term interest parity holds).
  - Domestic interbank market effectively reallocates reserves (no segmentation or shallow markets).

### D. Friction Cost, and Country and Counterparty Risk
- Introducing FX market frictions:
  - Spread between buying and selling rates and transaction costs implies FX flows respond only after domestic rates deviate from the foreign rate by at least +/- f.
  - Domestic short-term rates fluctuate in a corridor defined by i* +/- f.
  - In well-developed markets and absent country/counterparty risk, FX flows still offset reserve deviations once rates exceed friction costs, keeping reserves near equilibrium R.
- Introducing country and counterparty risk:
  - Local banks face a country-specific risk premium (Prem) that can be positive or negative and varies with fiscal authority credit rating; banking sector soundness; financial market depth; central bank liquidity management effectiveness; and FX reserve levels.
  - Risk premium can increase in times of stress.
  - The presence of Prem results in a neutral domestic interest rate different from the foreign interest rate: i = i* + Prem.
  - Capital-flow-based adjustment mechanism otherwise remains conceptually the same.

### E. Money and Foreign Exchange Market Dynamics
- Notation and markets:
  - Rd and Rs: demand and supply of bank reserves in the domestic money market; i: interest rate that equilibrates demand and supply.
  - FXd and FXs: demand and supply of FX. Demand for FX increases when domestic interest rate decreases below i* adjusted by Prem; supply of FX increases when domestic interest rate increases.
- Positive liquidity shock example (e.g., increase in government expenditures reducing government account at the central bank):
  - Domestic money market rate decreases from i to i2.
  - At i2, domestic rate is below i* + Prem; FXd at FX2d exceeds FXs at FX2s → excess demand for FX.
  - Central bank sells FX to banks to satisfy excess FX demand, reducing bank reserves supply back to initial Rs and restoring money market rate to initial level i; markets return to neutral.
- Increase in banks’ precautionary demand for reserves:
  - Rd increases to R2d while Rs unchanged → excess demand for reserves at new interest rate i2 (higher).
  - Higher domestic rates encourage banks to convert FX to local currency through the central bank; supply of reserves increases and short-term rates move back to i.
- Adverse terms-of-trade (ToT) shock:
  - Supply of FX falls while FX demand unchanged → excess demand for FX.
  - Central bank FX sales reduce supply of bank reserves below equilibrium R.
  - Domestic money market rate increases above i* + Prem, attracting capital inflows that offset initial ToT shock; money market rate, FX reserves, and bank reserves return to initial equilibrium, though with potentially large interim volatility depending on shock size.
- Overall mechanism summary:
  - Each fluctuation in autonomous factors can change domestic short-term money market rates, triggering capital flows and/or central bank FX intervention until re-equilibration.
  - Local short-term money market rate oscillates around the level implied by the UIP condition.

### Key findings, vulnerabilities, and dynamics
- Key relationships and equations preserved from the analysis:
  - FXI = R - R1
  - FXO = R2 - R
  - i = i* + Prem
  - FX flows triggered only when domestic rates deviate by more than +/- f from i* in presence of frictions.
- Vulnerabilities identified:
  - Friction costs, market segmentation, settlement delays, time-zone differences, and country/counterparty risk can impair seamless liquidity adjustment and delay or damp capital flows.
  - Higher volatility in short-term rates increases banks’ demand for precautionary reserves and reduces reliance on market funding, impairing money market development.
  - NFA can experience large swings under fixed exchange rate arrangements due to ToT shocks or sudden capital outflows, potentially producing sizeable liquidity shortfalls.
  - In extreme cases, persistent liquidity shortfalls may force domestic interest rates to remain high, causing illiquidity in the banking sector and risking solvency; conversely, persistent excess reserves deter money market development and the market’s ability to absorb idiosyncratic shocks.

*Source: the authors.*

### 11.1 Money Market                                                   11.2 FX Market

### 11.1 Money Market / 11.2 FX Market

### III. FROM THEORY TO PRACTICE — Overview
- Liquidity management under a fixed exchange rate with an open capital account has a dual objective: steering a short-term money market rate while defending the peg.
- Under some fixed arrangements where the central bank uses its policy rate to defend the peg (for example, Denmark), the policy rate offsets capital flows (exogenous shocks) while liquidity management instruments offset short-term autonomous factors’ shocks.
- Under flexible regimes, liquidity management aims primarily to steer a short-term money market rate that reflects the monetary policy stance, with usually limited financial stability considerations.
- In fixed exchange rate with open capital account, liquidity management often must also preserve financial stability by offsetting shocks.

### A. Liquidity Management Challenges under a Fixed Exchange Rate
- A purely FX-flow–based seamless liquidity management without central bank intervention can lead to significant volatility in short-term money market rates because of autonomous factors shocks and frictions; this justifies central bank intervention to stabilize short-term rates.
- In the absence of active liquidity management, volatile liquidity conditions could have the following consequences:
  - Higher precautionary demand for reserves, leading to an increase in funding costs and lending rates;
  - Shallow money markets due to persistent liquidity surpluses that reduce the need for interbank trading;
  - Unclear monetary policy signal when the central bank fails to stabilize short-term money market rates close to its policy rate;
  - Large and unexpected changes in short-term money market rates that challenge banks’ liquidity and solvency.
- Specific challenges for central banks under a fixed exchange rate with open capital account:
  - Ascertaining friction costs;
  - Determining the changes in the risk premium;
  - Assessing the precautionary demand for reserves as a function of the settlement risk and the level of development of the money markets;
  - Anticipating changes in the supply of bank reserves (that is, forecasting autonomous factors);
  - Determining the liquidity management instruments needed to offset liquidity shocks.

### B. Steering Short-Term Interest Rates under a Peg
- The level to which domestic interest rates should be steered depends on:
  - The level of domestic financial deepening;
  - Integration into global financial markets;
  - Intensity of capital flows.
- Highly globally integrated countries exposed to larger capital account shocks may need to offset persistent capital flows using their policy rates; less integrated countries may have more autonomy and may allow deviations from UIP.
- Sizeable and persistent deviations from the UIP induce imbalances in the money and FX markets and may increase the risk of speculation during reversals in liquidity conditions.
- Automatic short-term interest rate adjustments consistent with a fixed exchange rate can generate:
  - More short-term interest rate volatility;
  - Higher and unpredictable demand for precautionary reserves;
  - Lower money market deepening;
  - Potentially higher risk of speculation;
  - Overall less resilience to systemic and idiosyncratic liquidity shocks.
- Two main practical liquidity management approaches under fixed exchange rate arrangements:
  - First approach: central bank sets its policy rate in reference to the anchor country policy rate and adjusts the rate as the risk premium changes (UIP condition). The liquidity management framework resembles the standard one; the central bank should set up a rule to adjust its policy rate when necessary, sometimes to preserve external stability more than to close an inflation gap and an output gap. The central bank must address difficulties in calibrating OMOs amid volatile autonomous factor shocks.
  - Second approach: central bank lets the market determine short-term interest rates but uses liquidity management operations to dampen interest rate volatility. Interest rates are determined similarly to the conceptual framework, but part of expected automatic adjustment volatility is absorbed by liquidity management tools.
- International experience shows a wide range of approaches depending on macro-financial conditions and vulnerability to exogenous shocks (current account versus capital account shocks). Examples include frameworks with limited liquidity management (Bulgaria), passive liquidity management under structural liquidity surplus (Saudi Arabia), active liquidity management (Denmark), and a standard mid-corridor system (Morocco). Jordan and Qatar have an open capital account and an interest rate corridor in place.

### Operational guidance: setting the policy rate under a peg
- Where central banks choose to set the policy rate, it can be rule based and consistent with a UIP condition.
- The policy rate can be set close to the anchor country policy rate adjusted by a risk premium.
- Indicators and instruments for anchor and risk premium assessment:
  - O/N interest rate swaps for major international currencies provide an indication of the future interest rate path for those currencies.
  - Credit default swaps (CDSs) indicate pricing of sovereign risk; CDSs may not be available in every market and should be complemented by assessments of local banking sector credit risk and liquidity premium depending on market development.

### Country practices and examples
- Denmark (Danmarks Nationalbank, DNB):
  - DNB sets monetary policy interest rates under a pegged exchange rate to the euro: the lending rate; the certificates of deposit (CDs) rate; the discount rate; and the current-account rate.
  - DNB policy rates are geared to the management of the exchange rate and are not used for business cycle management.
  - When there are no pressures on the FX market, DNB usually changes its interest rates in step with the European Central Bank (ECB) policy rate.
  - The difference between DNB and ECB policy rates depends on relative market perception of country risk. The spread turned negative during the crisis as DNB tried to discourage capital inflows.
  - In a situation of negative interest rate on CDs, it is crucial that counterparties’ holdings of CDs are sufficient to ensure transmission from CDs’ interest rates to money-market rates.

- Hong Kong SAR (Hong Kong Monetary Authority, HKMA):
  - The base rate was introduced in September 1998 to address the tradeoff between interest rates’ responsiveness to capital flows and dampening excessive interest rate volatility.
  - The base rate is the interest rate used to compute the discount window rate.
  - The base rate is currently set at 50 basis points above the prevailing US Fed Fund target rate or the average of the five-day moving averages of the O/N and one-month HIBORs, whichever is the higher.
  - The HKMA announces the base rate every day before the interbank market opens in Hong Kong SAR.
  - Since 2009, the risk premium of the base rate (above the lower limit of the Fed Fund target) remained stable at 50 basis points.
  - Since December 2015, the base rate has followed the same tightening cycle as the Fed Fund target rates.
  - To maintain the peg, the HKMA purchased HKD and sold USD upon requests from banks at 7.85 Hong Kong dollars per US dollar, reducing aggregate balance in the banking system and raising interbank interest rates.

- Saudi Arabia (Saudi Arabian Monetary Authority, SAMA):
  - One key monetary policy rate, the O/N reverse repo rate, closely tracks the Fed Fund target (lower bound).
  - Monetary policy is decided by SAMA’s Monetary Policy Committee (MPC) that meets at a minimum eight times a year; ad hoc meetings may be called.
  - The Saudi Arabian banking system is in structural liquidity surplus mainly due to oil-related flows and the fixed exchange rate.
  - SAMA issues SAMA bills with one-, four-, 13-, 26-, and 52-week maturities to sterilize excess liquidity.
  - In response to a liquidity tightening in 2016, SAMA eased the loan-to-deposit ratio, reduced weekly issuance of SAMA bills from 9 billion SAR to 3 billion SAR, offered term repurchase facilities, and provided liquidity to banks through placements of deposits.
  - SAMA has a variable width interest rate corridor; since 2009 the system de facto operated as a floor due to liquidity surpluses. Since mid-2009, the floor (reverse repo rate) remained equal to the Fed Funds Target (lower bound).
  - The ceiling (O/N repo rate) stayed close to 2 percent from the beginning of 2009 until March 2018 when it was gradually increased to 3 percent currently.
  - The corridor narrowed with the Fed’s interest rate hikes since end-2015.

*Source: the authors.*

### 15.2 SAMA Policy Rates and SAIBOR

### 15.2 SAMA Policy Rates and SAIBOR

### Discount windows versus O/N standing facilities
- Central banks such as the DNB and the HKMA use a discount window rather than an O/N standing credit facility (SF) that imposes a firm ceiling on money market rates.
- Distinctions:
  - A discount window can be at longer maturities than O/N and allows more central bank flexibility.
  - An O/N SF is at the bank’s initiative, with a fixed interest rate and standard maturity; the central bank stands ready to supply whatever amount of O/N reserve balances are demanded at the fixed rate (the corridor ceiling).
- Operational constraints under fixed exchange rate arrangements:
  - Maintaining an O/N SF and fighting speculative attacks can be incompatible under rapid NFA declines and high devaluation expectations.
  - Example: the DNB abolished the O/N SF after several bouts of speculation and imposes limits on current-account deposits; excess deposits above overall limits are converted into CDs to prevent speculative buildup.
- Footnote: Under normal circumstances, HKMA discount window use is at a bank’s initiative based on prices and eligible collateral set by HKMA, rather than HKMA discretion.

### Interbank rates, excess liquidity, and volatility
- Following the Fed Funds target cut to 0–0.25 percent in 2009, interbank rates in Hong Kong SAR and Saudi Arabia stayed close to the zero lower bound.
- Large excess liquidity increases susceptibility to volatility when liquidity conditions reverse:
  - HIBOR and SAIBOR displayed increasing volatility in 2016–18.
- A fixed-width mid-corridor system can stabilize short-term money market rates when:
  - The economy is not subject to large capital inflows/outflows.
  - The local currency is not under unsustainable pressure.
- Preconditions for successful corridor under fixed exchange rate arrangements include:
  - adequate liquidity forecasting framework;
  - absence of speculation/unsustainable pressures on the peg;
  - existence of an active interbank market;
  - clear separation between monetary policy operations and lender-of-last-resort framework;
  - relatively developed fixed income markets providing sufficient eligible collateral.

### Open Market Operations (OMOs) and autonomous factor forecasts
- Many fixed-exchange-rate central banks use OMOs to offset short-term liquidity shocks and stabilize money market rates:
  - Jordan, Morocco, Nepal, and Oman conduct weekly repo OMOs.
  - Others offer term SFs to inject/withdraw liquidity.
- Weekly OMOs support market development; too frequent (daily) or too infrequent operations have drawbacks.
- Full-allotment auctions under a peg can defeat shock-offset objectives and risk over-injection.
- When in structural liquidity surplus, central banks issue auction-based short-term securities or deposits to absorb liquidity.
  - Bulgaria and Kuwait had a one-week auction-based deposit.
  - Nepal, Qatar, and Swaziland relied on bilateral deposits.
  - Central banks issuing securities include Bhutan, Denmark, Fiji, Hong Kong SAR, Jordan, Kuwait, Solomon Islands, Samoa, and UAE.
  - Few issue one-week tenor securities (Jordan, UAE); some issue up to five years.
- Issuing longer-term central bank securities can substitute for shallow government securities markets by replacing low-yield excess reserves with higher-yield, risk-free securities that lock up liquidity and increase HQLA for repos.
- Forecasting FX interventions under a peg is limited:
  - Short-term forecasts up to five days are feasible using t+2 settlement delay and aggregation of counterparties’ forecasts; reliable horizon limited to up to five days.
  - If settlement delay is t+0, FX interventions become unpredictable.
- When sizeable, persistent, and not properly forecasted, capital outflows and negative ToT shocks raise speculation risk and demand for precautionary reserves, which may be unstable.
- Limiting speculation risk often requires limiting banks’ excess reserves that could finance large FX purchases.

### Reserve requirements mechanism
- Averaged reserve requirements reduce the need for perfect liquidity forecasts by absorbing unexpected shocks and lowering demand for precautionary reserves when full averaging is allowed.
- Effectiveness requires enough averaging room over a relatively long reserve maintenance period.
- Reserve requirements can create a minimum domestic demand for central bank funding when the ratio generates a structural liquidity deficit.
- Tradeoff: Reserve requirement as buffer versus optimal stock of OMOs — low OMO demand and low central bank securities stock reduce OMO traction on short-term rates.
- Survey findings (ISIMP 2013 and MOID 2018):
  - Most currency boards/conventional pegs had reserve requirements with averaging, usually ~one month or more.
  - Eight countries/monetary unions had less than a four-week reserve averaging period:
    - ECCU, Lesotho, Nepal, Swaziland, UAE, and Venezuela (one week),
    - Bosnia Herzegovina (10 days),
    - Jordan (two weeks).
  - Denmark, Djibouti, Hong Kong SAR, and Kuwait had no reserve requirements.
  - Few countries remunerated required reserves.
- Some central banks prefer sterilizing surpluses via issuance of central bank instruments rather than reserve requirements; issuance of longer-term instruments cannot substitute the buffering role of reserve averaging.
- Required reserves should be remunerated close to their opportunity cost when ratio is relatively low and objective is not to curb excessive credit growth, to avoid altering counterparties’ funding costs relative to policy stance.
- Functional distinctions:
  - Reserve requirement ratio sterilizes large persistent liquidity shocks from exogenous shocks.
  - Reserve averaging absorbs unexpected short-term autonomous factor shocks.
  - Structural excess liquidity should be sterilized through structural instruments to leave room for interbank development and contain inflationary/credit pressures.
  - Reserve requirement ratio can be adjusted to mop up persistent FX reserve accumulation or reduced during persistent capital outflows/adverse current account shocks.
  - Averaging incentivizes interbank trading by allowing undershoot/overshoot on any day.

### Standing facilities (SFs)
- SFs include O/N credit and deposit facilities; spreads should exceed domestic money market transaction costs to encourage interbank trading and limit SF recourse.
- Challenges under fixed exchange rate arrangements:
  - A deposit rate set too high increases facility reserves, discouraging capital outflows and increasing central bank liquidity management cost.
  - Counterparties may use standing credit facility to purchase FX against domestic-currency collateral under peg pressure, increasing devaluation likelihood.
- Consequently, spreads between liquidity management facilities under pegs are usually large.
- Survey results: Few currency board/conventional peg central banks had both O/N standing credit and deposit facilities: Fiji, Jordan, Morocco, Qatar, Saudi Arabia, and Swaziland.

### Refinancing needs and collateral framework
- Aggregated refinancing needs can arise under pegs without capital controls for two main reasons:
  - Large persistent autonomous factor changes (e.g., ToT shock like oil price slump) draining reserves.
  - Large increase in demand for precautionary reserves due to heightened counterparty and credit risk perceptions, causing segmentation and some counterparties to refuse lending.
- Persistent negative liquidity shocks are first absorbed by ex ante buffers (central bank securities, reserve requirements); once depleted, central bank refinancing serves as the ultimate liquidity buffer.
- In thin interbank markets, idiosyncratic liquidity shocks frequently create refinancing demand; central bank may provide refinancing OMOs or credit facilities if eligible collateral exists; otherwise emergency liquidity assistance is needed.
- A relatively large stock of government securities is preferred collateral; where shallow, central banks should develop broader collateral frameworks, including credit claims with risk controls and haircuts, to equalize residual risk across collateral classes and support emergency liquidity assistance.

### Offsetting FX flows and offset coefficients — framework and equations
- This paper adopts an operational approach: assess central bank capacity to offset (sterilize) FX flows using policy rate, reserve requirements, and OMOs, deriving equations from central bank balance sheet equality (assets = liabilities).
- Disaggregation of central bank balance sheet used:
  - NFA = R + LA – LP + CiC + GOV + OIN
    - R = bank reserves at central bank
    - LA = central bank liquidity absorption operations
    - LP = central bank liquidity-providing operations
    - CiC = currency in circulation
    - GOV = government account at central bank
    - OIN = other items of the balance sheet
- Central bank control may extend only to bank reserves, not monetary base (CiC autonomous).
- Estimated equations (first differences) to assess offsetting behavior:
  - Equation (2): ∆NFA_i,t = α1 ∆i_i,t + α2 ∆R_i,t + α3 ∆LP_i,t + α4 ∆LA_i,t + α5 ∆CiC_i,t + α6 ∆GOV_i,t + ε_t
    - Test 1: H0 α1 = 0 (central bank does not use policy rate to offset NFA changes).
  - Equation (3): ∆LP_i,t = α1 ∆i_i,t + α2 ∆R_i,t + α3 ∆NFA_i,t + α4 ∆LA_i,t + α5 ∆CiC_i,t + α6 ∆GOV_i,t + ε_t
    - Test 2: H0 α3 = 0 (central bank does not adjust liquidity-providing OMO to sterilize NFA changes).
  - Equation (4): ∆LA_i,t = α1 ∆i_i,t + α2 ∆R_i,t + α3 ∆LP_i,t + α4 ∆NFA_i,t + α5 ∆CiC_i,t + α6 ∆GOV_i,t + ε_t
    - Test 3: H0 α4 = 0 (central bank does not adjust liquidity-absorbing OMO to sterilize NFA changes).
- In fixed exchange rate arrangements, NFA and bank reserves are expected to be more volatile absent central bank liquidity management; shallow domestic government bond markets (common with commodity export shocks) reduce available collateral.

### Instruments to offset NFA fluctuations
- Two main structural instruments:
  - Policy rate: useful when capital account shocks are responsive to policy rate; can stabilize both NFA and bank reserves.
  - Reserve requirements: needed when capital flows are unresponsive to policy rate and large FX reserve accumulation increases liquidity surpluses.
- The paper estimates country-specific quarterly regressions of equations (2), (3), and (4), including lags where needed.

### Denmark: empirical findings and operations
- Exchange rate and policy structure:
  - Denmark maintains a fixed exchange rate vis-à-vis the euro and participates in ERM at central rate 746.038 krone per 100 euro with a fluctuation band of +/- 2.25 percent.
  - DNB policy rates: lending rate, CDs rate, discount rate, current-account rate.
  - Monetary policy counterparties hold current accounts at DNB with O/N deposits accruing at current-account rate.
- Weekly OMOs:
  - On last banking day of each week, DNB conducts OMOs: counterparties can borrow vs collateral or deposit by purchasing DNB CDs, usually seven-day maturity.
  - Loans and deposits accrue at lending rate and CDs rate, respectively.
  - DNB fixes OMO rates; counterparties determine volumes.
  - On other banking days DNB offers purchase and sale of CDs and adds a premium in buybacks to incentivize market liquidity exchange rather than selling CDs to DNB.
- Liquidity adjusting operations (auctions or bilateral) can be deposits, lending against assets, or FX swaps, with counterparties varying in participation.
- DNB balance sheet stabilized since 2010 except a temporary spike in early 2015.
- Econometric estimation for Denmark:
  - Estimation period: 2000Q4–2017Q4; variables in first differences and stationary per ADF tests.
  - Using DNB lending rate as explanatory variable in equation (2): lending rate not statistically significant.
  - Using interest rate differential (DNB lending rate minus ECB policy rate, i-i*) as independent variable:
    - A 1 percentage point increase in the interest rate differential at t-3 increases ∆NFA by DKK 2.7 billion.
    - Coefficient statistically significant at 90 percent confidence (p-value < 0.1).
    - Reject H0 under Test 1: suggests interest rate differential may be used to offset changes in NFA.
  - Substituting ECB policy rate for the interest differential indicates DNB NFA are responsive to ECB policy rate with a negative relationship.
  - Equations (3) and (4) results:
    - Increase in ∆NFA results in decrease in DNB monetary policy lending (∆) and increase in CDs outstanding (∆).
    - Results statistically significant at p-value < 0.01, suggesting DNB sterilizes NFA effects on bank reserves by adjusting volumes of monetary operations.
- Reserve supply curve observations for Denmark:
  - Reserve supply curve was downward sloping during 2000–06 and 2007–14, possibly indicating NFA driving bank reserves and money market rates.
  - Reserve supply curve becomes upward sloping in 2015–17.

_Italic: Source: SAMA; content extracted from the supplied chapter wpiea2019058 - 15.2 SAMA Policy Rates and SAIBOR._

### 19.1 O/N Interbank Rate (Y axis) and Bank

### wpiea2019058 - 19.1 O/N Interbank Rate (Y axis) and Bank

### Denmark: Reserves Supply Curves and Interbank Rates (Figures 19–21)
- Figures shown (19.1–19.2, 20.1–20.2, 21.1–21.2) plot:
  - O/N interbank rate on the Y axis and Bank Reserves on the X axis for periods 2000–06, 2007–14, and 2015–17.
  - O/N interbank rate on the Y axis and NFA on the X axis for the same periods.
- Captioned source: DNB website and authors’ calculations.
- Note in the figure panel: "Only statistically significant variables are displayed."

### Hong Kong SAR: Currency Board Mechanics and Monetary Base
- Exchange rate arrangement:
  - HKD officially linked to the USD at the rate of 7.8 HKD per one USD.
  - Linked Exchange Rate System adopted on October 17, 1983.
  - Classified by IMF AREAER 2017 as de facto exchange rate anchor.
- Currency board implication:
  - Monetary base must be fully backed by foreign reserves (stock and flow).
  - Any change in the monetary base is fully matched by a corresponding change in foreign reserves at the fixed exchange rate.
  - Monetary base change described as a passive process (upon the request of banks).
  - Backing resources are kept in the Hong Kong SAR Exchange Fund.
- Components of the monetary base:
  - Certificates of Indebtedness (back banknotes issued by note-issuing banks).
  - Government-issued notes and coins in circulation.
  - Aggregate balance (sum of clearing account balances of banks kept with the HKMA).
  - Exchange Fund bills and notes (issued by the HKMA).
- Note-issuing banks:
  - When issuing banknotes, required to purchase Certificates of Indebtedness by submitting equivalent USD at HKD 7.80 per one US dollar to the HKMA for the Exchange Fund.
  - HKD banknotes are therefore fully backed by USD held by the Exchange Fund.
  - Redemption of Certificates of Indebtedness returns equivalent USD from the Exchange Fund.
- Interest rate adjustment channel:
  - Under the Currency Board, interest rates (not the exchange rate) adjust to inflows or outflows.
  - Monetary base increases when USD is sold to the Currency Board for HKD; contracts when USD is bought from the Currency Board.
  - Changes in the monetary base cause domestic interest rates to fall or rise, respectively, to counteract capital movements while exchange rate remains stable.
  - Process compared to the price-specie flow mechanism (referenced as described in Section II).
- Liquidity cushion and discount window:
  - Discount window provides O/N liquidity via repurchase agreements using Exchange Fund paper and other eligible securities as collateral.
  - The base rate forms the foundation upon which discount rates are computed.
  - HKMA provides discount window to reduce excessive interest rate volatility.

### Hong Kong SAR: Empirical Strategy and Results
- Data construction and estimation period:
  - Government-issued notes and coins in circulation combined with Certificates of Indebtedness to form CiC series.
  - Exchange Fund bills and notes used as proxy for liquidity absorption operations.
  - Discount window used as liquidity-providing operation.
  - Base rate used as policy rate.
  - Placements by fiscal reserves used as proxy for government account.
  - Hong Kong Exchange Fund FX reserves used as proxy for NFA.
  - Estimation period: 2001Q3–2017Q4.
- Stationarity and variables:
  - All variables in regressions are in first difference and stationary according to Augmented Dickey-Fuller tests, except DCIC and DGOV.
- Regression findings:
  - Equation (2) — relationship between base rate and NFA:
    - Regression coefficient associated to the base rate change (one period ahead) is statistically significant.
    - Discount window volume coefficient is not statistically significant.
    - Appendix VII, Table 1 displays regression results including only significant explanatory variables.
    - Note: inclusion of only one non-stationary variable (DCIC) limits risk of spurious regression (DGOV is not statistically significant).
    - HKMA commentary: "the significance of base rate does however not imply that it uses the base rate to react to FX reserves."
  - Equation (3) — discount window volume response to NFA:
    - None of the explanatory variables is statistically significant.
    - Interpretation: HKMA does not adjust the volume of the discount window to changes in NFA; discount window not calibrated according to autonomous factors forecasts.
  - Equation (4) — Exchange Fund bills and notes response to NFA:
    - Results confirm that an increase in the NFA results in an increase in the volume of Exchange Fund bills and notes issued.
    - Appendix VII, Table 2 includes explanatory variables that are statistically significant only.
    - Note: inclusion of only one non-stationary variable (DGOV) limits risk of spurious regression (DCIC is not statistically significant).
    - HKMA indicates it does not issue bills and notes to sterilize the effect of the NFA; additional supply of short-dated Exchange Fund paper is designed to meet strong demand by banks for liquidity management.
    - HKMA view: increase in Exchange Fund Bills represents a change in the composition of the monetary base (shift from aggregate balance to Exchange Fund paper); monetary base remains fully backed by foreign exchange reserves; issuance not expected to have a significant impact on liquidity conditions and interest rates.

### Conclusions and Policy Recommendations (Paper-wide)
- Theoretical and practical insights:
  - The paper presents a theoretical framework for liquidity management under a fixed exchange rate derived from the price-specie flow mechanism.
  - Under the framework, liquidity management is automatic and does not rely on central bank monetary operations.
  - The theoretical reserves supply curve is upward sloping in absence of central bank intervention; in practice it depends on FX flows and their responsiveness to the local money market rate.
  - Autonomous factor shocks and the reaction of FX flows to changes in liquidity conditions can generate excessive interest rate volatility.
  - Liquidity management that stabilizes short-term rates has a financial stability function beyond traditional monetary policy implementation in fixed exchange rate arrangements.
- Empirical cross-country findings on practices of central banks under currency boards and pegs:
  - Majority introduced reserve requirements, usually with an approximately one-month averaging period.
  - Many use short-term OMOs, but not always calibrated according to autonomous factor forecasts.
  - Few have an interest rate corridor in place.
  - In some cases, the interest rate is used to offset FX flows.
- Policy recommendations for central banks operating under fixed exchange rates and open capital accounts:
  - With adequate combination of instruments (reserve requirements, interest rate, OMOs, and SFs) and calibration of OMOs according to autonomous factors’ forecasts, central banks can:
    - Offset short-term liquidity shocks, dampen interest rate volatility, and foster money market stability and development.
    - Lower the precautionary demand for reserves, funding costs, and lending rates.
    - Build liquidity buffers in good times and contain the buildup of financial stability risks, while releasing liquidity in times of reversals in liquidity conditions (capital outflows or adverse current account shocks).

### Appendix highlights (selected mechanics and instruments)
- Interest rate setting mechanism (Appendix I) — selected entries:
  - Hong Kong SAR: central bank steering short-term rate? N. "The base rate is set with reference to the Fed Fund Target."
  - Denmark: steering short-term rate? N? "The lending rate is set with reference to the ECB policy rate."
  - Morocco: steering short-term rate? Y. "Combination of Taylor Rule and UIP condition. The central bank has a standard mid-corridor system."
- Fixed Exchange Rate Open Market Operations (Appendix II) — selected entries:
  - ECCU: Repo and Reverse Repo Y, Irregular frequency, 1 week maturity, Uniform price.
  - Jordan: Repo and Reverse Repo Y, Weekly, 1 week and 1 month, Fixed rate, partial allotment.
  - Hong Kong SAR: Repo and Reverse Repo N.
- FX Swap usage (Appendix II) — selected:
  - Bahrain: Y, Daily, 1 week; 1, 3, and 6 months, Fixed rate, full allotment.
  - UAE: Y, Daily, 1 week to 12 months, Fixed rate, full allotment.
  - Hong Kong SAR: entry blank/unclear.
- Collateralized lending (Appendix II) — selected:
  - Denmark: Y, 1 week, Fixed rate, full allotment.
  - Morocco: Y, 1 year, Fixed rate, full allotment.
  - Hong Kong SAR: N.
- Central bank securities (Appendix II) — selected:
  - Hong Kong SAR: Y, maturities 1, 3 to 6 months, 1 year and 2 years, Multiple price.
  - UAE: Y, maturities 1 week; 1, 2, 3, 6, 12 months; 2, 5 years, Fixed rate, full allotment.
- Standing facilities (Appendix III) — selected:
  - Hong Kong SAR: O/N SF Credit N, O/N SF Deposit N.
  - Jordan: O/N SF Credit Y, O/N SF Deposit Y.
  - Kuwait: O/N SF Credit Y, O/N SF Deposit N.
- Reserve requirements (Appendix IV) — selected entries (local currency reserve requirements only):
  - Bosnia and Herzegovina: Y, Ratio 10, Averaging Y, Averaging Period 10 days, Remuneration 0.
  - Brunei: Y, Ratio 6, Averaging Y, Averaging Period 1 month, Remuneration 0.
  - Bulgaria: Y, Ratio 10, Averaging Y, Averaging Period 1 month, Remuneration 0.
  - Hong Kong SAR: N (no local currency reserve requirement indicated).
  - Kuwait: N.
  - Morocco: Y, Ratio 4, Averaging Y, Averaging Period 4-5 weeks, Remuneration 0.75.
  - Nepal: Y, Ratio 5-6, Averaging Y, Averaging Period 7 days, Remuneration 0.
  - Qatar: Y, Ratio 4.75, Averaging N, Remuneration 0.

*Source: IMF staff compilation of the content unit wpiea2019058 (text and figures as provided).*

### Appendix V. Fixed Exchange Rate Direct Instruments of Monetary Control

### Appendix V. Fixed Exchange Rate Direct Instruments of Monetary Control

### Coverage of direct instruments by exchange rate arrangement and economy
- Table presents presence (“Y” = Yes; “N” = No; blank = information unclear or unavailable) of:
  - Interest rate Controls (*)
  - Credit Ceilings (*)
  - Directed Credits (*)
  - Specific Lending Requirements (*)
- Currency Boards (examples):
  - Bosnia and Herzegovina: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - Brunei: Interest rate Controls Y; Credit Ceilings Y; Directed Credits N; Specific Lending Requirements Y
  - Bulgaria: Interest rate Controls N; Credit Ceilings  N; Directed Credits N; Specific Lending Requirements N
  - Djibouti: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - ECCU: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - Hong Kong SAR: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
- Conventional Pegs (selected entries):
  - Aruba: (entries not provided / blank)
  - Bahamas: Interest rate Controls Y; Credit Ceilings Y; Directed Credits Y; Specific Lending Requirements Y
  - Bahrain: Interest rate Controls N; Credit Ceilings N; Directed Credits  N
  - Belize: Interest rate Controls Y; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - Bhutan: Interest rate Controls N; Credit Ceilings N; Directed Credits Y; Specific Lending Requirements Y
  - Cabo Verde: (entries not provided / blank)
  - CEMAC: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - Comoros: Interest rate Controls Y; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - Curaçao and Sint Maarten: (entries not provided / blank)
  - Denmark: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - Eritrea: (entries not provided / blank)
  - Fiji: Interest rate Controls N; Credit Ceilings N; Directed Credits Y; Specific Lending Requirements N
  - Iraq: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - Jordan: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - Kuwait: Interest rate Controls Y; Credit Ceilings Y; Directed Credits N; Specific Lending Requirements N
  - Lesotho: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - Libya: (entries not provided / blank)
  - Morocco: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - Namibia: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements Y
  - Nepal: Interest rate Controls N; Credit Ceilings Y; Directed Credits Y; Specific Lending Requirements Y
  - Oman: Interest rate Controls Y; Credit Ceilings Y; Directed Credits Y; Specific Lending Requirements Y
  - Qatar: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - Salomon Islands: Interest rate Controls Y; Credit Ceilings N; Directed Credits N; Specific Lending Requirements Y
  - Samoa: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - São Tomé and Principe: (entries not provided / blank)
  - Saudi Arabia: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements N
  - Swaziland: Interest rate Controls Y; Credit Ceilings N; Directed Credits N; Specific Lending Requirements Y
  - Turkmenistan: (entries not provided / blank)
  - UAE: Interest rate Controls N; Credit Ceilings N; Directed Credits N; Specific Lending Requirements Y
  - Venezuela: Interest rate Controls Y; Credit Ceilings Y; Directed Credits Y; Specific Lending Requirements Y
  - WAEMU: (entries not provided / blank)
- Note: (*) “Y” = Yes; “N” = No; “ ” = information unclear or unavailable in ISIMP and MOID; local currency reserve requirements only.
- Sources cited in the table: AREAER (2016), ISIMP (2013), MOID (2018) and central banks’ available information.

---

### Appendix VI. Denmark—Offset Coefficients Estimates

### Relationship between DNB NFA and Interest Rate Differential (Table 1)
- Dependent variable: Change in the NFA (DNFA)
- Coefficients (standard errors in parentheses):
  - Interest rate differential at t-3 (IDIFF(-3)): 2.695068*** (1.502025)
  - Change in Bank reserves (DR): 1.034366* (0.023093)
  - Change in DNB monetary policy lending (DLP): -1.038834* (0.022616)
  - Change in DNB CDs (DLA): 0.965646* (0.019960)
  - Change in CiC (DCIC): 0.934669* (0.282874)
  - Change in Government Deposits (DGOV): 0.970327* (0.017615)
- Goodness of fit:
  - R-squared 0.992165
  - Adjusted R-squared 0.991543
- Note: Standard errors in parentheses. * p < 0.01; ** p < 0.05; *** p < 0.1.

### Relationship between DNB NFA and ECB Policy Rate (Table 2)
- Dependent variable: Change in the NFA (DNFA)
- Coefficients (standard errors in parentheses):
  - Change in ECB policy rate at t-3 (DIF(-3)): -4.930738* (1.578297)
  - Change in Bank reserves (DR): 1.023118* (0.022426)
  - Change in DNB monetary policy lending (DLP): -1.030796* (0.022568)
  - Change in DNB CDs (DLA): 0.982617* (0.019520)
  - Change in CiC (DCIC): 0.976535* (0.272535)
  - Change in Government Deposits (DGOV): 0.978557* (0.016761)
- Goodness of fit:
  - R-squared 0.992886
  - Adjusted R-squared 0.992312
- Note: Standard errors in parentheses. * p < 0.01; ** p < 0.05; *** p < 0.1.

### Relationship between DNB Monetary Policy Lending and NFA (Table 3)
- Dependent variable: Change in DNB monetary policy lending (DLP)
- Coefficients (standard errors in parentheses):
  - Change in Bank reserves (DR): 0.957952* (0.031230)
  - Change in DNB NFA (DNFA): -0.930317* (0.019455)
  - Change in DNB CDs (DLA): 0.912844* (0.020558)
  - Change in CiC (DCIC): 0.947734* (0.260860)
  - Change in Government Deposits (DGOV): 0.914373*  (0.021014)
- Goodness of fit:
  - R-squared 0.976974
  - Adjusted R-squared 0.975579
- Note: Standard errors in parentheses. * p < 0.01; ** p < 0.05; *** p < 0.1.

### Relationship between DNB CDs Volume and NFA (Table 4)
- Dependent variable: Change in DNB CDs Volume (DLA)
- Coefficients (standard errors in parentheses):
  - Change in Bank reserves (DR): -1.026867* (0.036075)
  - Change in DNB monetary policy lending (DLP): 1.059994* (0.023872)
  - Change in DNB NFA (DNFA): 1.003091* (0.020532)
  - Change in CiC (DCIC): -0.964808* (0.284106)
  - Change in Government Deposits (DGOV): -0.992189* (0.017520)
- Goodness of fit:
  - R-squared 0.985630
  - Adjusted R-squared 0.984759
- Note: Standard errors in parentheses. * p < 0.01; ** p < 0.05; *** p < 0.1.

- Sources for Denmark estimates: ECB and DNB website and authors’ calculations.

---

### Appendix VII. Hong Kong SAR—Offset Coefficient Estimates

### Relationship between Hong Kong SAR Exchange Fund FX Reserves and HKMA Base Rate (Table 1)
- Dependent variable: Change in the NFA (DNFA)
- Coefficients (standard errors in parentheses):
  - Change in HKMA Base Rate at t-1 (DI(-1)): 15597.83** (6532.251)
  - Change in Bank reserves (DR): 0.964590* (0.101687)
  - Change in HKMA Exchange Fund Bills and Notes (DLA): 1.002436* (0.125564)
  - Change in CiC (DCIC): 4.078170* (0.501440)
- Goodness of fit:
  - R-squared 0.701477
  - Adjusted R-squared 0.687032
- Note: Dependent variable: Change in the NFA (DNFA). Standard errors in parentheses. * p < 0.01; ** p < 0.05; *** p < 0.1.
- Sources: HKMA website and authors’ calculations.

### Relationship between Hong Kong SAR Exchange Fund Bills and Notes and FX Reserves (Table 2)
- Dependent variable: Change in Exchange Fund Bills and Notes (DLA)
- Coefficients (standard errors in parentheses):
  - Change in Bank reserves (DR): -0.331331* (0.079627)
  - Change in HKMA NFA (DNFA): 0.418733* (0.043389)
  - Change in Government Deposits (DGOV): -0.159675** (0.064356)
- Goodness of fit:
  - R-squared 0.499424
  - Adjusted R-squared 0.483781
- Note: Dependent variable: Change in HKMA Certificates of Indebtedness volume (DLA). Standard errors in parentheses. * p < 0.01; ** p < 0.05; *** p < 0.1.

---

*Source: wpiea2019058 - Appendix V, Appendix VI, Appendix VII (tables and notes as provided).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019058.pdf_