## _wp12215

## Source details

**Canonical URL:** [_wp12215](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12215.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12215.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12215.pdf.json)

---

### I. INTRODUCTION
- Poland has based its monetary policy on inflation targeting (IT) since 1998.
- Current inflation target: 2.5 percent with a tolerance band for symmetrical deviations of one percentage point.
- The zloty began to freely float in 2000 after gradual abandonment of a currency peg to a composite of euro and U.S. dollar.
- The National Bank of Poland (NBP) implements monetary policy via open market operations (OMOs) aimed at keeping the POLONIA (transaction value-weighted overnight interbank rate) close to the NBP’s policy rate.
- Main instrument for OMOs: seven-day NBP bill issued every Friday; reference rate is the policy rate.
- Corridor for POLONIA determined by interest rates on Lombard credit and NBP’s deposit facility.
- In 2008 the operating target was changed from WIBOR SW to the overnight POLONIA.
- Paper analyzes determinants of the POLONIA spread (POLONIA minus policy rate) over January 2008–December 2011, covering the 2008 financial crisis and developments into the euro area crisis.
- Motivation: Since the start of the 2008–09 financial crisis, POLONIA has persistently been below the policy rate, suggesting limited influence of NBP OMOs on short-term interbank rate and potential impacts on monetary policy transmission. Also tests for external spillovers given Polish banking system linkages to global financial institutions.

### II. POLONIA BEHAVIOR AND VOLATILITY
- Empirical observations and stylized facts:
  - POLONIA follows tightening and easing cycles and moves within corridor set by standing deposit facility and Lombard credit.
  - Since 2009, POLONIA has often stayed below the policy rate persistently.
  - The mean of the spread can be characterized by an autoregressive process.
  - There are clear signs of volatility clustering with significant persistence in the squared spread.
  - Consistent with Kliber and Płuciennik (2011): post-Lehman, NBP retained influence over POLONIA but to a smaller degree than before.
- GARCH model diagnostics:
  - Long-run mean of the POLONIA spread = -0.32 percentage points.
  - Unconditional variance = 0.46.
  - Persistence of the variance = 0.92.
- Counts of days with positive spreads:
  - 10 days in 2009
  - 4 days in 2010
  - 22 days in 2011

### III. DETERMINANTS OF THE POLONIA SPREAD
- Conceptual framing:
  - Spread influenced by liquidity supply and demand in the interbank market, external factors, and interbank market structure.
  - NBP liquidity injection/withdrawal operations influence liquidity supply; banks’ refinancing needs determine liquidity demand.

- Liquidity supply — empirical facts and hypotheses:
  - Structural liquidity defined: NBP net foreign assets + net credit to government - currency in circulation + other items net.
  - Poland’s structural liquidity surplus has increased by 3.5 times since adoption of IT.
  - Increase more pronounced since 2009, mainly due to accumulation of net foreign assets (NFA) from NBP’s FX purchases associated with EU fund inflows and Eurobond issuance receipts from the Ministry of Finance (MOF).
  - To sterilize FX inflows, NBP expanded OMOs by issuing larger amounts of seven-day NBP bills each Friday.
  - NBP’s FX purchases from the MOF and the corresponding MOF’s zloty withdrawal from the NBP increase structural liquidity in the interbank market.
  - If NBP FX purchases are not accompanied by MOF withdrawal, structural liquidity does not increase because increase in NFA is offset by decline in NBP’s net credit to the government.
  - MOF converting EU funds on the FX spot market (since May 2011, at times) does not influence structural liquidity; abandoning this practice would increase structural liquidity.
  - Fine-tuning operations:
    - Two short-term fine-tuning operations introduced in December 2010; more in 2011.
    - Since June 2011, regular fine-tuning operations at end of reserve maintenance period by selling one-day NBP bills.
    - Hypothesis: these operations, particularly regular ones, should help to narrow negative spread and dampen its volatility.
  - Repo operations:
    - Initiated as part of "Confidence Pact" after Lehman collapse: six repo operations in 2008 with maturities six days to three months; six-month repo operations added in May 2009.
    - Three-month and six-month repo operations conducted once a month with pre-announced transaction dates.
    - Six-month repo operations discontinued in April 2010; three-month repo operations discontinued in October 2010. Currently no repo operations.
    - Expectation: repo operations would narrow a positive spread and dampen volatility; they may cause the spread to widen if the spread is negative.
  - Directional prediction on spread from NBP FX purchases: will cause the spread to widen if it is negative and to narrow if it is positive.
  - Influence on volatility from NBP FX purchases: difficult to predict.

- Liquidity demand — empirical facts and hypotheses:
  - Banks’ liquidity demand and participation in OMOs affect the POLONIA spread.
  - When market calm and NBP projects liquidity demand accurately, allotment of NBP bills is met by demand.
  - During liquidity problems or perceived interbank market problems, NBP faces difficulties projecting liquidity demand as demand for NBP bills is weaker.
  - In such circumstances banks prefer overnight liquidity management over locking funds in seven-day NBP bills.
  - After the collapse of Lehman Brothers, NBP stopped announcing allotment amount and let market demand determine amount; reverted to usual practice in February (contextual year implied).

- External factors:
  - Increasing stress in the euro interbank market (measured by EURIBOR-OIS changes) tightens liquidity in Poland’s interbank market and influences the spread.
  - When EURIBOR-OIS 3-month spread rose, the Polish interbank market became tighter.

- Interbank market structure:
  - The unsecured overnight market accounts for about 90 percent of trading; little activity beyond one week.
  - Poland’s interbank repo market is underdeveloped; repo turnover is about half the size of the unsecured interbank market and is small relative to the secondary market for treasury securities (average daily turnover of 24 billion zloty).
  - Factors inhibiting repo market development include lack of a widely-accepted master agreement, unclear terminology between repo and sell-buy-back/buy-sell-back, and tax regulation disadvantaging repo transactions.
  - Increased interbank credit risk (3-month WIBOR-OIS spread) is associated with a widening of the POLONIA spread when the spread is negative and a narrowing when positive.

### IV. EMPIRICAL MODEL AND DATA
- Sample and variables:
  - Daily observations from January 2008 to December 2011.
  - Total observations of the POLONIA spread: 1,014 (1,013 used after adjustments in estimation).
- Model specification:
  - Exponential GARCH (EGARCH) framework with a GARCH(3,4) specification chosen by a general-to-specific methodology.
  - Error term assumed to follow a Student’s t-distribution due to fat tails.
  - Exogenous dummies included: BUY_P, BUY_N, DRFT, DOFT, SREPO_P, DREPO_P, SREPO_N, DREPO_N, D/S<1, D/S>=1, D1–D8, EUROIS-EUROIS(-1), WIBOIS-WIBOIS(-1).

### V. EMPIRICAL RESULTS — MEAN EQUATION (selected)
- Persistence and distributional features:
  - SPREAD(-1): 0.972 (Std. Error 0.00713; z-Statistic 131.518; p value 0.000)
  - T-DIST. DOF: 2.625 (Std. Error 0.301; z-Statistic 8.720; p value 0.000)
- Significant mean-effect coefficients (coefficient, Std. Error, z-Statistic, p value):
  - BUY_N: -0.028 0.008 -3.335 0.001
  - DRFT_EX: 0.928 0.109 8.483 0.000
  - DOFT_UNEX: 0.076 0.031 2.432 0.015
  - SREPO_P: -0.222 0.076 -2.919 0.004
  - SREPO_P(-1): -0.298 0.164 -1.822 0.068
  - DREPO_P(-1): -0.305 0.047 -6.524 0.000
  - DREPO_N(-1): -0.032 0.015 -2.173 0.030
  - D/S<1: -0.106 0.013 -8.216 0.000
  - D1: -0.195 0.089 -2.205 0.027
  - D2: -0.057 0.017 -3.356 0.001
  - D4: -0.058 0.018 -3.268 0.001
  - D5: -0.042 0.017 -2.516 0.012
  - D7: -0.037 0.014 -2.568 0.010
  - EUROIS-EUROIS(-1): 0.145 0.074 1.964 0.050
  - WIBOIS-WIBOIS(-1): -0.096 0.028 -3.398 0.001

### VI. EMPIRICAL RESULTS — VARIANCE EQUATION (selected)
- GARCH/ARCH dynamics (coefficient, Std. Error, z-Statistic, p value):
  - C_var: -1.983 0.181 -10.967 0.000
  - ABS(RESID(-1)/SQRT(GARCH(-1))): 0.688 0.151 4.545 0.000
  - ABS(RESID(-2)/SQRT(GARCH(-2))): -0.246 0.128 -1.923 0.054
  - ABS(RESID(-3)/SQRT(GARCH(-3))): 0.300 0.109 2.755 0.006
  - LN(GARCH(-1)): 1.162 0.077 15.078 0.000
  - LN(GARCH(-2)): -0.920 0.123 -7.477 0.000
  - LN(GARCH(-3)): 0.532 0.114 4.657 0.000
  - LN(GARCH(-4)): -0.111 0.065 -1.719 0.086
- Exogenous impacts on variance (coefficient, Std. Error, z-Statistic, p value):
  - BUY_P: -0.531 0.181 -2.933 0.003
  - DRFT_EX: -2.055 0.681 -3.019 0.003
  - DOFT_UNEX: 0.866 0.384 2.255 0.024
  - SREPO_N(-1): 0.912 0.415 2.198 0.028
  - DREPO_N: 0.792 0.344 2.302 0.021
  - D/S>=1: 0.868 0.211 4.113 0.000
  - D/S<1: 1.158 0.180 6.431 0.000
  - D1: 3.784 0.390 9.693 0.000

### VII. MAIN EMPIRICAL CONCLUSIONS
- A persistently negative POLONIA spread since 2009 is observed, linked to structural liquidity in the banking system and banks’ reluctance to lock up liquidity in seven-day NBP bills.
- Banks’ frontloading of reserve requirements contributes to increasingly negative spreads and higher volatility toward the end of reserve maintenance periods.
- NBP fine-tuning operations, especially regular ones, are effective in narrowing negative spreads and reducing volatility.
- Repo operations helped steer POLONIA toward the policy rate but are associated with higher volatility when the spread is negative.
- External euro-area market sentiment affects spread levels; increases in EURIBOR-OIS are associated with tighter domestic liquidity.
- The interbank market’s underdevelopment (limited repo usage, overnight focus) constrains banks’ ability to meet short-term funding gaps and affects monetary transmission.

### VIII. POLICY IMPLICATIONS AND RECOMMENDATIONS
- Short term:
  - Provide the NBP with enough information to forecast liquidity accurately; enhance coordination between the MoF and the NBP so the NBP has advance information regarding MOF’s anticipated exchanges of foreign currency inflows into zloty.
  - Assess liquidity management instruments regularly and, if needed, introduce new instruments on a trial basis to gauge market demand and expectations (e.g., regular fine-tuning operations).
- Medium to long term:
  - Enhance interbank market activity and develop long-term local currency funding to improve banks’ asset and liability management and strengthen monetary policy transmission.
- Broader conditions:
  - External environment stability is important to reduce banks’ incentives to hoard liquidity; some influencing factors are beyond NBP control.

### IX. KEY SUMMARY STATISTICS FROM ESTIMATION
- Included observations: 1,013 after adjustments
- R-squared: 0.644
- Adjusted R-squared: 0.635
- S.E. of regression: 0.309
- Sum squared resid: 94.359
- Log likelihood: 374.157
- Durbin-Watson stat: 2.153
- Mean dependent var: -0.495
- S.D. dependent var: 0.512
- Akaike info criterion: -0.622
- Schwarz criterion: -0.336
- Hannan-Quinn criter.: -0.513

### X. APPENDIX I — STRUCTURAL LIQUIDITY
- Definition:
  - Structural liquidity refers to the aggregate liquidity position of the banking system, which corresponds to the sum of autonomous factors that are beyond the control of the central bank in the very short run.
- Simplified central bank balance sheet items (as listed in the source):
  - Assets: Net foreign assets; Net credit to government; Lending to banks/OMOs; Other items, net
  - Liabilities: Currency in circulation; Minimum reserve requirements (MRR); Excess reserves
- Liquidity supply identity (presented exactly):
  - Liquidity supply =   Net foreign assets  
    +   net credit to government    Autonomous factors  
    +   other items, net  
    -    currency in circulation  
    +   lending to banks/OMOs    Policy factor
- Structural liquidity positions and implications:
  - Structural liquidity surplus:
    - Occurs when the structural liquidity exceeds the MRR.
    - The banking system has a structural liquidity surplus with respect to the central bank.
    - The central bank may conduct OMOs to absorb surplus liquidity from the banking system.
    - Example from the text: Poland’s banking system is currently in a structural liquidity surplus position.
  - Structural liquidity deficit:
    - Occurs when the structural liquidity is exceeded by the MRR.
    - The banking system has an aggregate liquidity deficit with respect to the central bank.
    - Lending to banks by the central bank may be needed to relieve the liquidity shortage.
  - Balanced structural liquidity:
    - Structural liquidity is equal to the MRR.

*Source: _wp12215 (IMF staff estimates and analysis, January 2008–December 2011 period).*

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

### _wp12215 - References

### I. INTRODUCTION
- Poland has based its monetary policy on inflation targeting (IT) since 1998.
- Current inflation target: 2.5 percent with a tolerance band for symmetrical deviations of one percentage point.
- The zloty began to freely float in 2000 after gradual abandonment of a currency peg to a composite of euro and U.S. dollar.
- The National Bank of Poland (NBP) implements monetary policy via open market operations (OMOs) aimed at keeping the POLONIA (transaction value-weighted overnight interbank rate) close to the NBP’s policy rate.
- Main instrument for OMOs: seven-day NBP bill issued every Friday; reference rate is the policy rate.
- Corridor for POLONIA determined by interest rates on Lombard credit and NBP’s deposit facility.
- In 2008 the operating target was changed from WIBOR SW to the overnight POLONIA to reflect concentration of overnight interbank activity and POLONIA’s transaction-based liquidity price.
- Paper analyzes determinants of the POLONIA spread (POLONIA minus policy rate) over January 2008–December 2011, covering the 2008 financial crisis and developments into the euro area crisis.
- Motivation: Since start of the 2008–09 financial crisis, POLONIA has persistently been below the policy rate, suggesting limited influence of NBP OMOs on short-term interbank rate and potential impacts on monetary policy transmission. Also tests for external spillovers given Polish banking system linkages to global financial institutions.

### II. POLONIA AND ITS DEVIATION FROM THE POLICY RATE
- Empirical observations:
  - POLONIA follows tightening and easing cycles and moves within corridor set by standing deposit facility and Lombard credit.
  - Since 2009, POLONIA has often stayed below the policy rate persistently.
  - The mean of the spread can be characterized by an autoregressive process.
  - There are clear signs of volatility clustering with significant persistence in the squared spread.
  - Consistent with Kliber and Płuciennik (2011): post-Lehman, NBP retained influence over POLONIA but to a smaller degree than before.
- GARCH model results (following Panigirtzoglou and others (2000) specification):
  - Long-run mean of the POLONIA spread = -0.32 percentage points.
  - Unconditional variance = 0.46.
  - Persistence of the variance = 0.92, implying that following an episode of high volatility, volatility is expected to remain high.
- Additional noted counts:
  - Number of days with positive spreads: 10 days in 2009, 4 days in 2010, and 22 days in 2011.

### III. DETERMINANTS OF THE POLONIA SPREAD: PREDICTIONS
- Conceptual framing:
  - The POLONIA spread is influenced by interaction of liquidity supply and demand in the interbank market, external factors, and interbank market structure.
  - NBP liquidity injection/withdrawal operations influence liquidity supply; banks’ refinancing needs determine liquidity demand.

- A. Liquidity Supply — empirical facts and hypotheses
  - Structural liquidity defined: NBP net foreign assets + net credit to government - currency in circulation + other items net (see Appendix I in source).
  - Poland’s structural liquidity surplus has increased by 3.5 times since adoption of IT.
  - Increase more pronounced since 2009, mainly due to accumulation of net foreign assets (NFA) from NBP’s FX purchases associated with EU fund inflows and Eurobond issuance receipts from the Ministry of Finance (MOF).
  - To sterilize FX inflows, NBP expanded OMOs by issuing larger amounts of seven-day NBP bills each Friday.
  - The NBP’s FX purchases from the MOF and the corresponding MOF’s zloty withdrawal from the NBP increase structural liquidity in the interbank market.
  - If NBP FX purchases are not accompanied by MOF withdrawal, structural liquidity does not increase because increase in NFA is offset by decline in NBP’s net credit to the government.
  - MOF converting EU funds on the FX spot market (since May 2011, at times) does not influence structural liquidity; abandoning this practice would increase structural liquidity.
  - Fine-tuning operations:
    - Two short-term fine-tuning operations introduced in December 2010; more in 2011.
    - Since June 2011, regular fine-tuning operations at end of reserve maintenance period by selling one-day NBP bills.
    - Hypothesis: these operations, particularly regular ones, should help to narrow negative spread and dampen its volatility.
  - Repo operations:
    - Initiated as part of "Confidence Pact" after Lehman collapse: six repo operations in 2008 with maturities six days to three months; six-month repo operations added in May 2009.
    - Three-month and six-month repo operations conducted once a month with pre-announced transaction dates.
    - Six-month repo operations discontinued in April 2010; three-month repo operations discontinued in October 2010. Currently no repo operations.
    - Expectation: repo operations would narrow a positive spread and dampen volatility; they may cause the spread to widen if the spread is negative.
  - Directional prediction on spread from NBP FX purchases: will cause the spread to widen if it is negative and to narrow if it is positive.
  - Influence on volatility from NBP FX purchases: difficult to predict.

- B. Liquidity Demand — empirical facts and hypotheses
  - Banks’ liquidity demand and participation in OMOs affect the POLONIA spread.
  - When market calm and NBP projects liquidity demand accurately, allotment of NBP bills is met by demand.
  - During liquidity problems or perceived interbank market problems, NBP faces difficulties projecting liquidity demand as demand for NBP bills is weaker.
  - In such circumstances banks prefer overnight liquidity management over locking funds in seven-day NBP bills.
  - After the collapse of Lehman Brothers, NBP stopped announcing allotment amount and let market demand determine amount; reverted to usual practice in February (year implied in source context).

*Source: _wp12215 - References (excerpts from IMF staff estimates and analysis, January 2008–December 2011 period).*

### 2009. To a lesser degree, during 2011, banks preferred to have a liquidity cushion and a

### _wp12215 - 2009. To a lesser degree, during 2011, banks preferred to have a liquidity cushion and a

### Determinants of the POLONIA spread
- Liquidity supply
  - NBP’s purchases of FX from the MOF widen the spread when the spread is negative; impact is insignificant when spread is positive.
  - Regular fine-tuning operations narrow a negative spread and reduce volatility; non-regular fine-tuning operations have a smaller narrowing effect on level and tend to increase volatility.
  - Repo operations reduce the spread when the spread is positive; when the spread is negative, repo impact on the spread is relatively small. Repo operations are associated with higher volatility, particularly when the spread is negative.
- Liquidity demand
  - Underbidding (Demand/Supply < 1) is associated with a widening of the spread if the spread is negative and a narrowing if the spread is positive.
  - Frontloading of reserve requirements: banks hold excess reserves at the start of the maintenance period and release them toward the end, producing an increasingly negative spread and higher volatility toward the end of the maintenance period. Most day dummies for the last days of the reserve maintenance period have negative coefficients (D1–D8), particularly Day 1.
- External factors
  - Increasing stress in the euro interbank market (measured by EURIBOR-OIS changes) tightens liquidity in Poland’s interbank market and influences the spread.
  - When EURIBOR-OIS 3-month spread rose, the Polish interbank market became tighter.
- Interbank market structure
  - The unsecured overnight market accounts for about 90 percent of trading; little activity beyond one week.
  - Poland’s interbank repo market is underdeveloped; repo turnover is about half the size of the unsecured interbank market and is small relative to the secondary market for treasury securities (average daily turnover of 24 billion zloty).
  - Factors inhibiting repo market development include lack of a widely-accepted master agreement, unclear terminology between repo and sell-buy-back/buy-sell-back, and tax regulation disadvantaging repo transactions.
  - Increased interbank credit risk (3-month WIBOR-OIS spread) is associated with a widening of the POLONIA spread when the spread is negative and a narrowing when positive, consistent with banks hoarding liquidity.

### Empirical model and data
- Sample and variable scope
  - Daily observations from January 2008 to December 2011.
  - Total observations of the POLONIA spread: 1,014 (1,013 used after adjustments in estimation).
- Model specification
  - Exponential GARCH (EGARCH) framework with a GARCH(3,4) specification chosen by a general-to-specific methodology.
  - Error term assumed to follow a Student’s t-distribution due to fat tails.
  - Exogenous dummies included: NBP FX purchase (Buy_P, Buy_N), regular fine-tuning (DRFT), ad-hoc fine-tuning (DOFT), repo start/settlement dummies split by sign (SREPO_P, DREPO_P, SREPO_N, DREPO_N), auction underbidding/overbidding (D/S<1, D/S>=1), last 8 days of reserve maintenance (D1–D8), EURIBOR-OIS change (EUROIS-EUROIS(-1)), WIBOR-OIS change (WIBOIS-WIBOIS(-1)).
  - Dummies for the last days of maintenance period included from Day 1 to Day 8 based on statistical significance.

### Empirical results — mean equation (selected)
- Persistence and distributional features
  - Lagged spread coefficient SPREAD(-1): 0.972 (Std. Error 0.00713; z-Statistic 131.518; p value 0.000) — strong persistence.
  - Estimated degrees of freedom for the t-distribution: 2.625 (T-DIST. DOF 2.625; Std. Error 0.301; z-Statistic 8.720; p value 0.000) — confirms fat tails.
- Significant mean effects (coefficients, Std. Error, z-Statistic, p value)
  - BUY_N: -0.028 0.008 -3.335 0.001
  - DRFT_EX: 0.928 0.109 8.483 0.000
  - DOFT_UNEX: 0.076 0.031 2.432 0.015
  - SREPO_P: -0.222 0.076 -2.919 0.004
  - SREPO_P(-1): -0.298 0.164 -1.822 0.068
  - DREPO_P(-1): -0.305 0.047 -6.524 0.000
  - DREPO_N(-1): -0.032 0.015 -2.173 0.030
  - D/S<1: -0.106 0.013 -8.216 0.000
  - D1: -0.195 0.089 -2.205 0.027
  - D2: -0.057 0.017 -3.356 0.001
  - D4: -0.058 0.018 -3.268 0.001
  - D5: -0.042 0.017 -2.516 0.012
  - D7: -0.037 0.014 -2.568 0.010
  - EUROIS-EUROIS(-1): 0.145 0.074 1.964 0.050
  - WIBOIS-WIBOIS(-1): -0.096 0.028 -3.398 0.001

### Empirical results — variance equation (selected)
- GARCH/ARCH dynamics (coefficients, Std. Error, z-Statistic, p value)
  - C_var: -1.983 0.181 -10.967 0.000
  - ABS(RESID(-1)/SQRT(GARCH(-1))): 0.688 0.151 4.545 0.000
  - ABS(RESID(-2)/SQRT(GARCH(-2))): -0.246 0.128 -1.923 0.054
  - ABS(RESID(-3)/SQRT(GARCH(-3))): 0.300 0.109 2.755 0.006
  - LN(GARCH(-1)): 1.162 0.077 15.078 0.000
  - LN(GARCH(-2)): -0.920 0.123 -7.477 0.000
  - LN(GARCH(-3)): 0.532 0.114 4.657 0.000
  - LN(GARCH(-4)): -0.111 0.065 -1.719 0.086
- Exogenous impacts on variance (coefficients, Std. Error, z-Statistic, p value)
  - BUY_P: -0.531 0.181 -2.933 0.003 (reduces volatility when spread positive)
  - DRFT_EX: -2.055 0.681 -3.019 0.003 (regular fine-tuning decreases volatility)
  - DOFT_UNEX: 0.866 0.384 2.255 0.024 (non-regular fine-tuning increases volatility)
  - SREPO_N(-1): 0.912 0.415 2.198 0.028
  - DREPO_N: 0.792 0.344 2.302 0.021
  - D/S>=1: 0.868 0.211 4.113 0.000 (overbidding increases volatility)
  - D/S<1: 1.158 0.180 6.431 0.000 (underbidding increases volatility)
  - D1: 3.784 0.390 9.693 0.000 (last day of maintenance period strongly increases volatility)

### Summary of main empirical conclusions
- A persistently negative POLONIA spread since 2009 is observed, linked to structural liquidity in the banking system and banks’ reluctance to lock up liquidity in seven-day NBP bills.
- Banks’ frontloading of reserve requirements contributes to increasingly negative spreads and higher volatility toward the end of reserve maintenance periods.
- NBP fine-tuning operations, especially regular ones, are effective in narrowing negative spreads and reducing volatility.
- Repo operations helped steer POLONIA toward the policy rate but are associated with higher volatility when the spread is negative.
- External euro-area market sentiment affects spread levels; increases in EURIBOR-OIS are associated with tighter domestic liquidity.
- The interbank market’s underdevelopment (limited repo usage, overnight focus) constrains banks’ ability to meet short-term funding gaps and affects monetary transmission.

### Policy implications and recommendations
- Short term
  - Provide the NBP with enough information to forecast liquidity accurately; enhance coordination between the MoF and the NBP so the NBP has advance information regarding MOF’s anticipated exchanges of foreign currency inflows into zloty.
  - Assess liquidity management instruments regularly and, if needed, introduce new instruments on a trial basis to gauge market demand and expectations (e.g., regular fine-tuning operations).
- Medium to long term
  - Enhance interbank market activity and develop long-term local currency funding to improve banks’ asset and liability management and strengthen monetary policy transmission.
- Broader conditions
  - External environment stability is important to reduce banks’ incentives to hoard liquidity; some influencing factors are beyond NBP control.

### Key summary statistics from estimation
- Included observations: 1,013 after adjustments
- R-squared: 0.644
- Adjusted R-squared: 0.635
- S.E. of regression: 0.309
- Sum squared resid: 94.359
- Log likelihood: 374.157
- Durbin-Watson stat: 2.153
- Mean dependent var: -0.495
- S.D. dependent var: 0.512
- Akaike info criterion: -0.622
- Schwarz criterion: -0.336
- Hannan-Quinn criter.: -0.513

*Source: IMF staff estimates and analysis contained in the provided PDF content.*

### APPENDIX I. STRUCTURAL LIQUIDITY

### APPENDIX I. STRUCTURAL LIQUIDITY

### Definition and concept
- Structural liquidity refers to the aggregate liquidity position of the banking system, which corresponds to the sum of autonomous factors that are beyond the control of the central bank in the very short run.
- The supply of liquidity through autonomous factors can be derived from a simplified balance sheet of the central bank.

### A Simplified Balance Sheet of the Central Bank
- Assets
  - Net foreign assets
  - Net credit to government
  - Lending to banks/OMOs
  - Other items, net
- Liabilities
  - Currency in circulation
  - Minimum reserve requirements (MRR)
  - Excess reserves

### Liquidity supply decomposition
- The appendix presents the liquidity supply identity exactly as shown:
  - Liquidity supply =   Net foreign assets  
    +   net credit to government    Autonomous factors  
    +   other items, net  
    -    currency in circulation  
    +   lending to banks/OMOs    Policy factor

### Structural liquidity positions and implications
- Three situations related to structural liquidity are possible:
  - Structural liquidity surplus
    - Occurs when the structural liquidity exceeds the MRR.
    - The banking system has a structural liquidity surplus with respect to the central bank, meaning that it does not need to obtain funding from the central bank.
    - The central bank may conduct OMOs to absorb surplus liquidity from the banking system.
    - Example from the text: Poland’s banking system is currently in a structural liquidity surplus position.
  - Structural liquidity deficit
    - Occurs when the structural liquidity is exceeded by the MRR.
    - The banking system has an aggregate liquidity deficit with respect to the central bank.
    - Lending to banks by the central bank may be needed to relieve the liquidity shortage.
  - Balanced structural liquidity
    - Structural liquidity is equal to the MRR.

*APPENDIX I. STRUCTURAL LIQUIDITY*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12215.pdf_