## MOLDOVA’S INFLATION TARGETING REGIME

## Source details

**Canonical URL:** [MOLDOVA’S INFLATION TARGETING REGIME](https://www.imf.org/-/media/files/publications/selected-issues-papers/2026/english/sipea2026025.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/selected-issues-papers/2026/english/sipea2026025.pdf.md)
- [Structured JSON version](/-/media/files/publications/selected-issues-papers/2026/english/sipea2026025.pdf.json)

---

### A. Context
- Adoption and framework:
  - Formal inflation targeting (IT) regime adopted in late 2012 with operationalization in 2013.
  - Amended NBM Law: price stability as primary objective with an inflation target range of 5+/- 1.5 percent.
  - Transparency and forward-looking policy: quarterly Inflation Reports with macroeconomic forecasts, risk assessments, and policy rationale; forecast-based decision-making supported by a macroeconomic model and scenario analysis; improved communication via regular press releases and policy statements.
- Inflation dynamics and pass-through:
  - Inflation remained close to the 5 percent target during the first two years of the IT regime but has been very volatile since end-2014.
  - Large share of goods in the CPI (food, fuels) and regulated prices (gas, electricity) have led to exogenous volatility, including occasional large adjustments related to Russian gas supply and pricing up to 2025.
  - High share of imported goods in the CPI contributed to sizable exchange rate pass-through to consumer prices.
  - VAR estimates (technical annex A):
    - median pass-through to CPI for MDL-US dollar exchange rate is 45 percent;
    - for MDL-euro exchange rate the pass-through elasticity is 53 percent.
    - For exchange rate changes driven by domestic monetary policy shocks the pass-through from an exchange rate change to the CPI is 62 percent for both the MDL-US dollar rate and the MDL-euro rate.
    - Interpretation: around half of an exchange rate movement transmits to the CPI for those shocks.
- Post-2014 banking crisis:
  - Banking crisis following bank fraud in 2014 caused capital flight and a strong depreciation in 2015.

### B. The NBM’s Evolving Monetary Policy Operations and Implications
- Instruments and operating environment:
  - Main operating target: base rate, complemented by reserve requirements (RRs), open market operations, and standing facilities.
  - Contextual challenges: fear of floating due to liability dollarization/high pass-through and shallow financial markets limiting transmission.
  - Excess liquidity after the 2014 bank fraud and emergency lending led the NBM to increase RRs to exceptionally high levels and to issue short-term certificates of deposit (CDs).
- Taylor rule evidence (technical annex B):
  - Sample split: pre-2020 and post-2020 (rules guiding FXI adopted in late-2019).
  - 2013Q1-2019Q4: base rate did not significantly react to the inflation gap; instead, base rate increased by 7 basis points per 1 percent quarterly depreciation in the leu-dollar exchange rate (significant at 1 percent).
  - 2020Q1-2025Q2: base rate displays a significant reaction to the inflation gap; exchange rate coefficient smaller and significant only at the 10 percent level.
  - Table of Taylor Rule Coefficients (as presented):
    - 2013Q1-2025Q2: Inflation gap 0.1-0.03; Output gap 0-0.15-0.03; Change in exchange rate 0.07***0.07*** 0.04*
    - Notes: *0.05<p≤0.1, ** 0.01<p≤0.05, *** p≤0.01.
- Interest rate pass-through and transmission (technical annex C):
  - Transmission has strengthened over the past decade.
  - Since 2021:
    - Deposit rate rises 0.19 percentage points on impact after a one-percentage point increase in the base rate.
    - Pass-through complete after about 4 months with a total increase of 0.55 percentage points.
    - Lending rate rises less: after 3 months it settles at 0.3 percentage points higher than before the base rate hike.
    - T-bill rates: 71 percent of base rate hikes were transmitted to T-bills within a few months.
  - Recent episode (energy price hike in January 2025):
    - Base rate increase of 290 bps was followed by an increase of bank loan rates of 110 bps until July 2025;
    - Estimated pass-through to lending was 0.38 in this episode.
  - Interpretation: lower pass-through to lending than deposits likely reflects strong borrower-lender relationships and less elastic deposits.
- FX interventions (FXI) and reaction function (technical annex D):
  - Use of FXI has declined substantially in recent years.
  - Pre-2020: a one percent month-over-month depreciation of the leu against the dollar associated with FX sales of 0.09 percent of annual GDP.
  - Post-2020: reaction coefficient not significant.
  - Interventions have become smaller relative to GDP and less biased toward dollar sales; smaller footprint in FX market relative to FX interbank transactions.
  - Degree of exchange rate management, measured as ratio of FXI volatility over the sum of FXI and exchange rate volatility, has decreased.
  - NBM did not target the leu-euro rate with FXI even as euro gained importance; replacing the dollar with the euro in the FXI reaction function shows no significant reaction to the euro rate.
- Reserve requirements and costs (technical annex E):
  - RRs raised on deposits in lei from 14 to 35 percent between February and November 2015 and to 42.5 percent in October 2018 to reduce excess liquidity and support transmission.
  - Required reserves remuneration:
    - first five percent not remunerated;
    - remaining 95 percent remunerated at the overnight deposit rate, currently two percentage points below the base rate.
  - Distortion measured as opportunity cost of total required reserves:
    - in December 2025, with RR reduced to 20 percent, the cost stood at about MDL 487 mln, roughly 1.6 percent of banks’ equity.
- Model simulation on RR reduction (IPF model, Adrian et al, 2021):
  - Experiment: 10 percent reduction in the RR from 22 to 12 percent; reduction perfectly credible and anticipated; adjustment gradual over four years with half-way point at 18 months.
  - Banking variables: nominal loan rates and spreads fall on announcement; stocks of loans and deposits rise; market-to-book ratios rise.
  - Macroeconomic effects: small in the model; caveat—new credit directed to households boosting asset prices and consumption in short term; benefits from boosting business lending not captured.
- CDs and repos (bank-specific liquidity tools):
  - CDs used intensively over past decade to shed excess liquidity beyond RRs; repo rate set at +0.25 pp above base rate; maturities generally 14 days.
  - CDs peaked just below 5 percent of GDP in March 2018 and again in July 2023; by December 2025 CDs declined to 1.4 percent of GDP.
  - Repos used less overall but picked up in late 2025; simultaneous sales of CDs and use of repos imply NBM absorbed liquidity from some banks while providing liquidity to others, substituting for interbank market activity.

### C. Policy Discussion and Recommendations
- Positive developments:
  - Greater focus on the inflation gap in setting the base rate strengthens IT practice and bolsters NBM credibility and accountability.
  - FXI practices improved: rules set in 2019 limit FXI to addressing FX market disruptions and not to stem trend exchange rate movements.
  - Reduced and less systematic use of FXI allows exchange rate to reflect market forces more.
- Use cases for FXI (IPF/IMF references summarized in text):
  - i) to address destabilizing premia from arbitrage frictions in shallow FX markets;
  - ii) to counter financial stability risks from FX mismatches;
  - iii) to support price stability when exchange rate changes risk de-anchoring inflation expectations.
  - Relevance to Moldova: shallow FX markets and 20-percent share of FX loans in total loans imply cases i) and ii) remain pertinent; macroprudential tools are first-line defense against currency mismatches.
- Recommendations on reserve requirements:
  - NBM should reduce RRs further once inflationary pressures have subsided.
  - Rationale:
    - Excess liquidity substantially reduced as credit to GDP ratio increased from 21 to 29 percent since early-2024.
    - Credit growth slowed toward end-2025 but projected to continue, further reducing excess liquidity.
  - Suggested approach:
    - Pursue a carefully calibrated reduction of RRs while closely monitoring inflation developments.
    - Consider a slow release over time to avoid undesirable credit booms; absorb any remaining excess liquidity through issuance of CDs if needed.
    - Vary speed of RR reductions depending on credit demand cycles to manage induced credit expansion risks.
  - Endpoint calibration:
    - Regional peers' local currency RRs range currently between 5.5 and 10 percent (presented country list and maturities as in source).
    - Optimal RR endpoint should consider banks’ liquidity needs given shock profile and degree of interbank market development.
- Financial market development implications:
  - Lower RRs would support government debt market development and interbank market revival:
    - Transition from liquidity absorber to liquidity provider would enable greater use of repos for open market operations using government debt as collateral.
    - Higher demand for government debt would raise valuations and support sovereign bond market development.
    - Scarcity of liquidity and a functioning sovereign bond market should help revive the interbank market, aided by ending fixed-rate-full-allotment and improved liquidity management and forecasting by banks and the NBM.
- Institutional recommendations to maintain credibility:
  - Continue building track record of fighting inflation and adherence to IT principles.
  - Follow recommendations in the Central Bank Transparency Code Review to ensure clear communication of deliberations and decisions.
  - Strengthen NBM’s political, institutional, and financial independence.

*Source: “MOLDOVA’S INFLATION TARGETING REGIME,” Selected Issues Paper (February 10, 2026), Republic of Moldova — International Monetary Fund.*

### References_______________________________________________________________________________ 9

### MOLDOVA’S INFLATION TARGETING REGIME

### A. Context
- Moldova adopted a formal inflation targeting (IT) regime in late 2012 with operationalization in 2013.
- The amended NBM Law enshrined price stability as the primary objective with an inflation target range of 5+/- 1.5 percent.
- To enhance transparency and forward-looking policy, the NBM:
  - began publishing quarterly Inflation Reports with macroeconomic forecasts, risk assessments, and policy rationale;
  - introduced forecast-based decision-making supported by a newly developed macroeconomic model and scenario analysis;
  - improved communication via regular press releases and policy statements.
- Inflation dynamics and pass-through:
  - Inflation remained close to the 5 percent target during the first two years of the IT regime but has been very volatile since end-2014.
  - Large share of goods in the CPI (food, fuels) and regulated prices (gas, electricity) have led to exogenous volatility, including occasional large adjustments related to Russian gas supply and pricing up to 2025.
  - High share of imported goods in the CPI contributed to sizable exchange rate pass-through to consumer prices.
  - VAR estimates (technical annex A): median pass-through to CPI for MDL-US dollar exchange rate is 45 percent; for MDL-euro exchange rate the pass-through elasticity is 53 percent.
  - For exchange rate changes driven by domestic monetary policy shocks the pass-through from an exchange rate change to the CPI is 62 percent for both the MDL-US dollar rate and the MDL-euro rate.
  - Interpretation: around half of an exchange rate movement transmits to the CPI for those shocks.
- Post-2014 banking crisis episode:
  - A banking crisis following bank fraud in 2014 caused capital flight and a strong depreciation in 2015.

### B. The NBM’s Evolving Monetary Policy Operations and Implications
- Instruments and environment:
  - The NBM uses the base rate as the main operating target, complemented by reserve requirements (RRs), open market operations, and standing facilities.
  - Challenges in a low-income country context include fear of floating due to liability dollarization/high pass-through and shallow financial markets limiting transmission.
  - Excess liquidity after the 2014 bank fraud and emergency lending led the NBM to increase RRs to exceptionally high levels and to issue short-term certificates of deposit (CDs).
- Taylor rule evidence (technical annex B):
  - Sample split: pre-2020 and post-2020 (rules guiding FXI adopted in late-2019).
  - 2013Q1-2019Q4: base rate did not significantly react to the inflation gap; instead, base rate increased by 7 basis points per 1 percent quarterly depreciation in the leu-dollar exchange rate (significant at 1 percent).
  - 2020Q1-2025Q2: base rate displays a significant reaction to the inflation gap; exchange rate coefficient smaller and significant only at the 10 percent level.
  - Table of Taylor Rule Coefficients (as presented):
    - 2013Q1-2025Q2: Inflation gap 0.1-0.03; Output gap 0-0.15-0.03; Change in exchange rate 0.07***0.07*** 0.04*
    - Notes: *0.05<p≤0.1, ** 0.01<p≤0.05, *** p≤0.01.
- Interest rate pass-through and transmission (technical annex C):
  - Transmission has strengthened over the past decade.
  - Since 2021:
    - Deposit rate rises 0.19 percentage points on impact after a one-percentage point increase in the base rate.
    - Pass-through complete after about 4 months with a total increase of 0.55 percentage points.
    - Lending rate rises less: after 3 months it settles at 0.3 percentage points higher than before the base rate hike.
    - T-bill rates: 71 percent of base rate hikes were transmitted to T-bills within a few months.
  - Recent episode: energy price hike in January 2025 — base rate increase of 290 bps was followed by an increase of bank loan rates of 110 bps until July 2025; estimated pass-through to lending was 0.38 in this episode.
  - Interpretation: lower pass-through to lending than deposits likely reflects strong borrower-lender relationships and less elastic deposits.
- FX interventions (FXI) and reaction function (technical annex D):
  - Use of FXI has declined substantially in recent years.
  - Pre-2020: a one percent month-over-month depreciation of the leu against the dollar associated with FX sales of 0.09 percent of annual GDP.
  - Post-2020: reaction coefficient not significant.
  - Interventions have become smaller relative to GDP and less biased toward dollar sales; smaller footprint in FX market relative to FX interbank transactions.
  - Degree of exchange rate management, measured as ratio of FXI volatility over the sum of FXI and exchange rate volatility, has decreased.
  - NBM did not target the leu-euro rate with FXI even as euro gained importance; replacing the dollar with the euro in the FXI reaction function shows no significant reaction to the euro rate.
- Reserve requirements and costs (technical annex E):
  - RRs were raised on deposits in lei from 14 to 35 percent between February and November 2015 and to 42.5 percent in October 2018 to reduce excess liquidity and support transmission.
  - Required reserves are only partially remunerated: first five percent not remunerated; remaining 95 percent remunerated at the overnight deposit rate, currently two percentage points below the base rate.
  - Distortion measured as opportunity cost of total required reserves: in December 2025, with RR reduced to 20 percent, the cost stood at about MDL 487 mln, roughly 1.6 percent of banks’ equity.
- Model simulation on RR reduction (IPF model, Adrian et al, 2021):
  - Experiment: 10 percent reduction in the RR from 22 to 12 percent; reduction perfectly credible and anticipated; adjustment gradual over four years with half-way point at 18 months.
  - Banking variables: nominal loan rates and spreads fall on announcement; stocks of loans and deposits rise; market-to-book ratios rise.
  - Macroeconomic effects: small in the model; caveat—new credit directed to households boosting asset prices and consumption in short term; benefits from boosting business lending not captured.
- CDs and repos (bank-specific liquidity tools):
  - CDs use intense over past decade to shed excess liquidity beyond RRs; repo rate set at +0.25 pp above base rate; maturities generally 14 days.
  - CDs peaked just below 5 percent of GDP in March 2018 and again in July 2023; by December 2025 CDs declined to 1.4 percent of GDP.
  - Repos used less overall but picked up in late 2025; simultaneous sales of CDs and use of repos imply NBM absorbed liquidity from some banks while providing liquidity to others, substituting for interbank market activity.

### C. Policy Discussion and Recommendations
- Positive developments:
  - Greater focus on the inflation gap in setting the base rate is a welcome step toward normal IT practice and bolsters NBM credibility and accountability.
  - FXI practices improved: rules set in 2019 limit FXI to addressing FX market disruptions and not to stem trend exchange rate movements.
  - The NBM’s reduced and less systematic use of FXI allows exchange rate to reflect market forces more.
- Use cases for FXI (IPF/IMF references summarized in text):
  - i) to address destabilizing premia from arbitrage frictions in shallow FX markets;
  - ii) to counter financial stability risks from FX mismatches;
  - iii) to support price stability when exchange rate changes risk de-anchoring inflation expectations.
  - Relevance to Moldova: shallow FX markets and 20-percent share of FX loans in total loans imply cases i) and ii) remain pertinent; macroprudential tools are first-line defense against currency mismatches.
- Recommendations on reserve requirements:
  - The NBM should reduce RRs further once inflationary pressures have subsided.
  - Rationale:
    - Excess liquidity substantially reduced as credit to GDP ratio increased from 21 to 29 percent since early-2024.
    - Credit growth slowed toward end-2025 but projected to continue, further reducing excess liquidity.
  - Suggested approach:
    - Pursue a carefully calibrated reduction of RRs while closely monitoring inflation developments.
    - Consider a slow release over time to avoid undesirable credit booms; absorb any remaining excess liquidity through issuance of CDs if needed.
    - Vary speed of RR reductions depending on credit demand cycles to manage induced credit expansion risks.
  - Endpoint calibration:
    - Regional peers' local currency RRs range currently between 5.5 and 10 percent (presented country list and maturities as in source).
    - Determination of optimal RR endpoint should consider banks’ liquidity needs given shock profile and degree of interbank market development.
- Financial market development implications:
  - Lower RRs would support government debt market development and interbank market revival:
    - Transition from liquidity absorber to liquidity provider would enable greater use of repos for open market operations using government debt as collateral.
    - Higher demand for government debt would raise valuations and support sovereign bond market development.
    - Scarcity of liquidity and a functioning sovereign bond market should help revive the interbank market, aided by ending fixed-rate-full-allotment and improved liquidity management and forecasting by banks and the NBM.
- Institutional recommendations to maintain credibility:
  - Continue building track record of fighting inflation and adherence to IT principles.
  - Follow recommendations in the Central Bank Transparency Code Review to ensure clear communication of deliberations and decisions.
  - Strengthen NBM’s political, institutional, and financial independence.

*Source: “MOLDOVA’S INFLATION TARGETING REGIME,” Selected Issues Paper (February 10, 2026), Republic of Moldova — International Monetary Fund.*

### References

### References and Appendix I. Methodology

### References
- Adler, G., Chang, K. S., and Z. Wang (2020). Patterns of Foreign Exchange Intervention under Inflation Targeting. IMF WP/20/69.
- Adler, G., Chang, Mano, R., K. S., Shao, Y. (2025). Foreign Exchange Intervention: A Dataset of Public Data and Proxies. Journal of Money, Credit and Banking, Volume 57 (5), pp. 1241-1273.
- Adrian, T., Erceg, C.J., Kolasa, M., Linde, J. and Zabczyk, P. (2021). A Quantitative Microfoundd Model for the Integrated Policy Framework. IMF WP 2021/292.
- Engler, P., Giucci, R. (2015). The Transmission Mechanism in Moldova. Reasons for its Weakness and Recommendations for its Strengthening. GET Moldova Policy paper 1/2015.
- Forbes, K., Hjortsoe I., Nenova, T. (2018). The shocks matter: Improving our estimates of exchange rate pass-through. Journal of International Economics 114 (2018) 255–275.
- Gabaix, X., and Maggiore, M. (2015). International Liquidity and Exchange Rate Dynamics. The Quarterly Journal of Economics (2015), 1369–1420.
- German Economic Team Moldova (2015). Newsletter No. 29, May-June 2015.
- Gertler, M., and Karadi, P. (2011). A model of unconventional monetary policy. Journal of Monetary Economics, Vol. 58(1), 17-34.
- Iacoviello, M. (2005). House Prices, Borrowing Constraints, and Monetary Policy in the Business Cycle. American Economic Review, Vol. 95(3), 739–764.
- International Monetary Fund (2022). Challenges of IT Implementation in LLMICS: A Case Study of Moldova. Selected Issues Paper. IMF Country Report No 22/61.
- International Monetary Fund (2023). Integrated Policy Framework – Principles for the Use of Foreign Exchange Intervention. Washington, DC.
- Mishra, P., Montiel, P.J., and Spilimbergo, A. (2010). Monetary transmission in low income countries. IMF WP 10/223.
- Poghosyan, T. (2020). Unveiling the Effects of Foreign Exchange Interventions: Evidence from the Kyrgyz Republic, IMF WP 20/2019.
- Rojas, D., Vegh, C., Vuletin, G. (2022). The macroeconomic effects of macroprudential policy: Evidence from a narrative approach. Journal of International Economics 139.

### Appendix I. Methodology — Overview
- Covers methodology for estimating exchange rate pass-through (ERPT), Taylor rule estimation, transmission of the base rate to bank rates, the FXI reaction function, costs of reserve requirements, and effects of changes in the required reserve ratio.
- Data frequencies: monthly data for 2006-2025 (ERPT), quarterly for Taylor rule (subsamples 2013Q1-2019Q4 and 2020Q1-2025Q2), monthly for bank rate pass-through and FXI reaction (FXI split between 2013-2019 and 2020-2025M7 subsamples).

### A. Pass-Through Estimation
- Method:
  - A flat prior BVAR is estimated on monthly data for 2006-2025 to estimate ERPT from changes in the nominal exchange rate (NER), denoted as ΔS, to the CPI.
  - Identification via standard sign restrictions (Forbes et al., 2018) for monetary policy and exchange rate shocks.
  - ERPT is calculated based on VAR impulse-response functions x_h:
    - EERPT_T^k(d) = [C_E_I_h(d) / S_1(d)] ε_k (for posterior draws d = 1, ... , D, horizon h = 1...12 months, structural shock ε_k).
- Sign restrictions table (as presented):
  - Monetary policy shock: Base rate > 0; CPI^ < 0; Euro NER * < 0; Dollar NER* < 0; Free reserves < 0; FXI n/a.
  - Exchange rate shock: Base rate > 0; CPI^ > 0; Euro NER * > 0; Dollar NER* > 0; Free reserves n/a; FXI < 0.
  - Note: * A decrease denotes an appreciation of the Moldovan leu vs. the named currency; ** A decrease denotes sales of reserves (purchases of Lei); ^ CPI restriction holds at the 12-month horizon for y/y inflation; shaded fields: Forbes et al. (2018) restrictions (originally for quarterly data, so imposed for 3 months).
- Additional identification restriction:
  - A reasonableness prior limits ERPT up to 150 percent to avoid implausibly high ERPT estimates.
- Outputs:
  - Distributions for domestic monetary policy and exchange rate shocks after one year are shown in the source (figures referenced).

### B. Taylor Rule Estimation
- A Taylor-type rule with the base rate as the NBM’s policy instrument is estimated quarterly for two subsamples of the IT regime:
  - 2013Q1-2019Q4
  - 2020Q1-2025Q2
- Regression specification:
  - Includes a constant and two lags of the base rate.
  - Inflation gap = year-over-year CPI inflation rate minus the target rate of 5 percent.
  - Output gap proxied by industrial production, detrended using a Hodrick-Prescott filter.
  - Exchange rate change = quarter-over-quarter change in the MDL-dollar exchange rate.

### C. The Transmission of the Base Rate to Bank Rates
- Method:
  - Approach of Mishra et al. (2010) applied.
  - Estimated autoregressive distributed lag model with monthly data:
    - Δe_t = α Δe_{t−1} + β Δe_{t−2} + γ Δy_t + δ Δy_{t−1} + ε Δy_{t−2} + μ_t
    - e_t: endogenous variable of interest (lending rate, deposit rate, T-bill rate).
    - y_t: base rate.
- Effects:
  - Short term (ST) effect = coefficient γ.
  - Long-run (LT) effect = (γ+δ+ε)/(1−α−β).

### D. The FXI Reaction Function
- Specification:
  - FXI (expressed as a percentage of GDP) regressed on one lag of FXI, the m/m percentage change in the MDL-dollar exchange rate, and the inflation gap (y/y CPI inflation rate minus the target rate of 5 percent).
- Data and sample:
  - Monthly data.
  - Sample split between 2013-2019 and 2020-2025M7 subsamples.
  - FXI series source: Adler et al. (2020).

### E. Costs of Reserve Requirements
- Definitions:
  - Unit opportunity cost = difference between the policy rate and remuneration of reserves; determines the wedge between lending and deposit rates.
  - NBM pays interest at the overnight deposit rate (ONDR) on required reserves exceeding 5 percent of the reserve base (RB); this remuneration is currently 2 percentage points below the base rate.
  - First 5 percent of RB are not remunerated.
- Unit cost expression:
  - Unit cost = [5% * RB * Base rate + (RR ratio − 5%)* RB*(Base rate − ONDR)] / (RR ratio * RB)
  - Simplified: (5%/RR ratio)*base rate + [(RR ratio − 5%)/RR ratio]*(base rate − ONDR)
- Implications:
  - As the gap between the base rate and the ONDR changes rarely, the unit cost varies mostly with the base rate and the RR ratio.
  - Costs calculated for the banking system as a whole.
- Example numeric value:
  - At the required reserve ratio of 20 percent of the reserve base and a base rate of 5 percent in December 2025, the unit cost was 2.8 percent.
  - Total cost = unit cost × required reserves holdings.

### F. Effects of Changes of the Required Reserve Ratio
- Modeling framework:
  - Adapted from the Quantitative Integrated Policy Framework (QIPF, Adrian et al. 2021).
  - QIPF: two-country open economy New Keynesian model with trade linkages and frictions in the foreign exchange market producing a home country premium (Gabaix and Maggiore, 2015).
  - Extended model includes financial intermediaries that make loans and take deposits in local currency.
- Key model features:
  - Prices of domestic, imported, and exported goods sticky in local currency.
  - Monetary, fiscal, and foreign exchange policy tools available.
  - Households: savers and borrowers (equal-sized blocks); borrowers have higher rate of time preference (Iacoviello, 2005).
  - Housing market: borrowers face a loan-to-value (LTV) limit; asset price fluctuations affect credit demand and consumption.
  - Credit supply constrained by banks' ability to raise funding from depositors.
  - Banking moral hazard (as in Gertler and Kiyotaki (2010) and Gertler and Karadi (2011)) limits leverage and induces interest rate spreads as incentive payments.
  - Loans are for one period and carry an ex ante nominal fixed interest rate.
  - Banks accumulate net worth from retained profits; loan supply proportional to net worth.
- Role of reserves and RR:
  - Banks must hold minimum reserves at the central bank depending on the reserve requirement ratio (RR).
  - In the model, reserves have no use other than satisfying the reserve requirement and are remunerated at the central bank’s policy rate.
  - Deposits are also remunerated at the policy rate; hence no direct “tax on banks” from reserves.
  - Reserves occupy balance sheet “space”: holding reserves raises banks’ overall leverage (loans plus reserves to net worth), tightening the incentive constraint and leading to higher spreads and reduced lending capacity.
  - Key assumption: creditors view reserves as better collateral than loans (loans are more opaque, non-traded, and may require monitoring), so the tightness of the bank constraint depends on the mix of loans and reserves and overall leverage.

*Source: sipea2026025 - References and Appendix I. Methodology (IMF content provided).*

---


_Source: https://www.imf.org/-/media/files/publications/selected-issues-papers/2026/english/sipea2026025.pdf_