## mfdnthcea

## Source details

**Canonical URL:** [mfdnthcea](https://www.imf.org/-/media/files/publications/dp/2022/english/mfdnthcea.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/dp/2022/english/mfdnthcea.pdf.md)
- [Structured JSON version](/-/media/files/publications/dp/2022/english/mfdnthcea.pdf.json)

---

### Executive Summary — Context, Motivation, and High-Level Conclusions
- Context and motivation:
  - Over the past decade and a half the world economy confronted two major crises—the global financial crisis (GFC) and the COVID-19 pandemic.
  - In both crises central banks cut interest rates and deployed unconventional monetary policy tools; many countries were pushed into a liquidity trap with interest rates close to zero while public debt rose to historic highs.
  - By the end of 2020, policy rates were below 1 percent in more than 60 percent of the global economy.
  - Monetary finance (MF) is defined as the financing of the government via a permanent increase in the monetary base beyond the level consistent with the inflation target and without paying interest on the newly created monetary base.
- Central debate:
  - Whether central banks should reconsider the prohibition on MF given liquidity-trap episodes and sovereign stress.
- Empirical strategy summary:
  - Two complementary approaches:
    1. Large panel analysis linking money growth (monetary base) to subsequent inflation using data back to the 1950s.
    2. Examination of whether unconventional monetary policy (UMP) announcements during March 2020–December 2020 affected inflation expectations, with focus on EMDEs.
- High-level empirical conclusions:
  - Relation between monetary base growth and inflation depends on initial inflation, central bank independence, and fiscal position.
  - In low-inflation, high-independence, healthy-fiscal-position environments, monetary base increases are followed by modest price increases.
  - Under high initial inflation, weak central bank independence, or large fiscal deficits, increases in the monetary base are followed by considerable price increases.
  - Significant non-linearities: inflationary pressures increase more than proportionally with the size of monetary expansion.
  - No systematic evidence that 2020 UMP announcements raised inflation expectations in the EMDE sample studied; caveat that programs were modest and seen as one-off responses to an exceptional shock.
- Policy implications (summary):
  - The taboo against MF helped establish central bank independence and a barrier to fiscal dominance.
  - Limited circumstances may exist where MF could be beneficial (persistent liquidity traps or to prevent self-fulfilling sovereign debt crises).
  - Any experimentation with MF should:
    - Remain modest in size.
    - Be limited to countries with low inflation and sustainable fiscal positions.
    - Be used only in exceptional circumstances (persistent liquidity trap or risk of self-fulfilling sovereign debt crisis).
    - Be decided independently by central banks to enhance macroeconomic stability.
  - Historical evidence warns MF under fiscal dominance can have devastating consequences; extreme caution warranted.

### Theoretical Underpinnings of Monetary Finance
- Distinction and comparison with Quantitative Easing (QE):
  - QE
    - Goals: Macroeconomic stimulus
    - Effects on Central Bank Balance Sheets: Large temporary expansion
    - Departure from Inflation Targeting?: No
    - Main Risks: Central bank losses
  - MF
    - Goals: Macroeconomic stimulus; Prevent self-fulfilling crises
    - Effects on Central Bank Balance Sheets: Modest permanent expansion; Potentially large expansion off-equilibrium
    - Departure from Inflation Targeting?: Yes; Only off-equilibrium
    - Main Risks: Fiscal dominance; Central bank losses, fiscal dominance
- MF for macroeconomic stimulus — mechanics and key theoretical points:
  - Core mechanism: permanent increase in monetary base passed to public via fiscal actions (tax cuts, spending) funded by central bank transfers or by central bank holding government bonds indefinitely—analogy to Friedman's “helicopter” money.
  - MF can overcome Ricardian equivalence because money:
    - does not pay interest, reducing government interest bill relative to bond-financed stimulus;
    - is irredeemable—the government never repurchases outstanding money.
  - Remuneration of reserves matters: MF delivers interest savings only if additional reserves are not remunerated; if reserves are remunerated, MF may simply swap government bonds for equally remunerated reserves and fail to deliver fiscal savings.
  - In a liquidity trap MF can lower real rates via higher expected future inflation; outside a liquidity trap committing to permanent monetary base increase implies temporary inflation above target.
  - Credibility is crucial: if agents expect reversal, MF effects resemble QE; if MF is perceived as open-ended, risks include de-anchored inflation expectations and fiscal dominance.
- MF as sovereign backstop to prevent self-fulfilling crises:
  - Mechanism: credible central bank commitment to purchase distressed sovereign bonds by expanding monetary base can shift markets from a bad equilibrium (run) to a good equilibrium (low yields, sustainable debt).
  - Credibility requires some tolerance for inflation (central bank must be prepared to tolerate sufficient inflation if needed).
  - Sterilized alternatives exist (purchases funded without increasing monetary base), but may not be viable if investors fear de facto defaults or fiscal claims.
  - Risks: central bank losses if fundamentals justify default; fiscal dominance if government pressure forces continuous MF; required inflation to avert default may be very large (model evidence suggests possibly double-digit inflation for prolonged periods in some cases).

### Historical Evidence on the Association Between Money and Inflation
- Empirical approach:
  - Flexible local projections (Jordà (2005)) linking d logMit (money growth) to πit+h (inflation h years ahead) controlling for up to l = 10 lags of money growth and inflation, real GDP growth d logYit (up to l = 10 lags), and country and year fixed effects κi + τt.
  - Annual panel data from the 1950s to 2020 for up to 157 countries.
  - Two aggregates: Monetary base (MB) and M2.
- Key empirical findings (preserve numeric magnitudes exactly):
  - Impact of a 10 percent monetary increase:
    - A 10 percent increase in MB is associated with an increase in inflation by about 1.5 percentage points on impact and in the subsequent year.
    - A 10 percent increase in M2 is associated with an increase in inflation by about 2 percentage points (point estimates).
    - After 10 years, a 10 percent increase in MB and M2 is associated with an increase in the price level by about 6 and 8 percent, respectively.
    - The 90 percent confidence bands for M2 are much wider than for MB.
  - Nonlinearity by size of monetary expansion:
    - A 1 percent increase in MB is associated with a 0.3 percent increase in the price level after 10 years.
    - A 100 percent increase in MB is associated with an increase in the price level up to 94 percent after 10 years.
    - Implication: inflationary risks may increase more than proportionally with the size of monetary expansion.
  - Dependence on initial inflation:
    - Interaction of money growth with one-year lagged inflation is positive and statistically significant on impact.
    - Examples:
      - If initial inflation is 100 percent, a 10 percent increase in MB is associated with an additional increase in inflation by 9 percentage points.
      - If initial inflation is 2 percent, a 10 percent increase in MB is associated with at most a 0.9 percentage point increase in inflation one year ahead.
  - Role of central bank independence:
    - Interaction between money growth and the de jure index of central bank independence (Garriga 2016) is negative and statistically significant.
    - A low level of independence (25th percentile) implies an increase in inflation three times as large as for a high independence level (75th percentile) after a given MB increase.
    - Implication: a highly independent central bank can sustain an increase in the monetary base three times as large as a low-independence central bank while experiencing the same price pressures.
  - Role of fiscal balance:
    - Interaction of money growth with a dummy for fiscal deficit > 3 percent of GDP is positive and significant.
    - When fiscal deficit is below 3 percent there is no statistically significant association between MB and inflation.
    - When fiscal deficit exceeds 3 percent—with an average of 6.2 percent in the regression sample—an increase in MB is associated with a significant and persistent increase in inflation.
- Robustness:
  - Results robust to dropping observations with annual inflation above 100 percent.
  - Results unchanged when removing QE episodes in advanced economies.
  - Interactions of money growth with initial inflation, central bank credibility, and fiscal deficit jointly included yield similar results.
  - M2 shows similar qualitative influences but with less precise estimates.

### UMP Announcements During COVID-19 and Inflation Expectations (EMDE-focused empirical assessment)
- Sample and data:
  - UMPs in 49 advanced economies and EMDEs between March 2020 and December 2020.
  - Sample includes 15 cases of direct government financing (DGF) and 64 other UMP cases.
  - Real-time inflation expectations: Consensus Economics (monthly; 31 countries); WEO forecast vintages (quarterly; 18 countries).
  - Median program size referenced in analysis: 2% of GDP.
- Descriptive statistics (Annex 6):
  - Structural fiscal deficit in the sample ranges from –8.9 to 2.2 percent of GDP.
  - Average 10-year past inflation in the sample ranges from 0.8 to 21.4 percent.
  - Bank of Mauritius one-off transfer: equal to almost 15 percent of GDP; end-of-year inflation in 2020 closed at 2.7 percent.
- Main empirical observations on expectations:
  - Average inflation forecasts for 2020 before and after central bank announcements are stable for both DGF and other UMP programs.
  - No notable difference in inflation expectations between relatively small and relatively large UMP programs.
  - Cross-section and panel regressions do not find UMP-related variables statistically significant in most specifications.
- Sterilization and monetary base changes:
  - Announced UMPs translated nearly one-for-one into changes of the monetary base on average, though variation across countries exists.
  - Some countries partially sterilized UMPs (e.g., issuance of central bank bills), partially unwinding monetary base impacts.
  - Robustness tests that control for sterilization do not change the relation between UMPs and inflation expectations relative to baseline.
- Panel evidence specifics:
  - Baseline panel (monthly Consensus Economics inflation expectations) includes country and monthly time fixed effects.
  - April 2020 dummy is large, significant, and negative (global nosedive in inflation expectations due to COVID-19).
  - Lagged dependent variable significant; UMP-related variables not significant.
  - Alternative specifications: interaction between EM dummy and UMP size statistically significant in one specification but economically small—"a UMP announcement equal to 1 percent of GDP raises inflation expectations by 0.04 percentage points".
  - Effect concentrated on QE programs; DGF dummy among EMs not statistically significant (and has a negative coefficient).

### Annex 6 — EMDE Sample Heterogeneity, Historical Cases, and Policy Guidance
- Sample heterogeneity:
  - Structural fiscal deficit range: –8.9 to 2.2 percent of GDP.
  - Average 10-year past inflation range: 0.8 to 21.4 percent.
  - Median program size: 2% of GDP.
  - Notable outlier: Bank of Mauritius transfer ≈ 15 percent of GDP.
- Historical cautionary cases (Box 1):
  - Zimbabwe, 2007–08:
    - Peak month-over-month inflation rate: 7.96 x 10^10 percent.
    - Prices doubled every 24.7 hours at peak.
    - Budget deficits in 2005–08: 25–45 percent of GDP.
    - Outcomes: 100 trillion Zimbabwe-dollar note issuance; per capita GDP fell to less than half its level a decade earlier; 70 percent of the population underfed.
  - German hyperinflation, 1922–23: peaked at 29,500 percent (monthly).
  - Hungarian hyperinflation, 1945–46: peaked at 4.19 x 10^16 percent (monthly).
  - Suriname, 2015–2020:
    - Central bank extended loans equal to 11.6 percent of GDP in 2015; inflation rose from 3.9 percent in 2014 to 25.1 percent in 2015 and 52.4 percent in 2016.
    - Fiscal deficit grew from 6 percent of GDP in 2018 to 21.2 percent in 2019.
    - Central bank provided loans worth 15.9 percent of GDP to the government until July 2020.
    - IMF WEO forecasted 108.1 percent end-of-period inflation for 2020; actual 2020 outturn moderated to 61 percent.
    - Real GDP contracted 13.5 percent in 2020.
- Policy recommendations (Annex conclusions and guidance):
  - UMP announcements during COVID-19, including DGF in EMDEs, did not lead to an economically and statistically significant increase in inflation expectations in the examined sample and specifications.
  - The lack of systemic effect likely reflects modest program sizes, one-off nature, and clear exceptional shock context.
  - Cross-country long-run evidence indicates:
    - Increases in monetary base associated with modest price pressures only when:
      - initial inflation is low;
      - central bank independence/credibility is strong;
      - fiscal deficits are modest (below 3 percent of GDP).
    - High initial inflation, weak central bank independence, and large fiscal deficits amplify price responses and risk fiscal dominance.
    - Inflationary pressures increase more than proportionally with the size of the monetary expansion.
  - Recommended policy stance:
    - Further analysis of conditions for MF could be merited, but relaxing the taboo carries considerable risks.
    - Any experimentation with MF should be:
      - Limited to extreme circumstances.
      - Modest in size, given documented non-linearities.
      - Considered only by countries with low inflation and sustainable fiscal positions.
      - Conducted with central bank full independence to decide if and when MF is appropriate.
    - In contexts of elevated inflation, unsustainable fiscal positions, and weak central bank independence, policymakers should vigorously refrain from using MF due to high likelihood of triggering strong inflationary pressures with disastrous outcomes.

### Annex 1 — Monetary Finance in Historical Perspective (selected analytical history)
- The Great Depression:
  - Monetarist view (Friedman and Schwartz 1963): recession severity linked to insufficient money supply; stronger increase in monetary base could have offset deflationary pressures.
  - Friedman’s helicopter-money analogy illustrates money-financed tax cut or permanent monetary base increase as stimulus in a liquidity trap.
- Post-WWII and Great Moderation:
  - Time-inconsistency and inflation bias literature (Kydland and Prescott 1977; Barro and Gordon 1983) motivated stronger central bank independence and inflation targeting to avoid MF-induced inflation bias.
- Japan’s liquidity trap and renewed debate:
  - Krugman (1998): central banks should "credibly promise to be irresponsible" by committing to a permanent increase in money supply to raise expected inflation and lower real rates.
  - Eggertsson and Woodford (2003): formal models show committing to a lower interest rate path (higher money supply) after exiting a trap can stimulate demand.
  - Matching monetary increases with fiscal stimulus strengthens permanence of expansion by altering incentives for future monetization.
- Key analytical insights:
  - MF can change the present discounted value (PDV) of taxes via:
    - Interest-rate savings if money pays lower interest than government bonds.
    - Irredeemable property of money if lim_{j → ∞} Λ_{j,t} M_{j+1} > 0, allowing permanent increases in money stock to enable lower taxation.
  - Degree of price stickiness determines output versus price effects of MF.

### Annex Table 7.2 — Regression Evidence on 2021 Inflation Forecast after UMP Announcement (specifications 1–7)
- Dependent variable: First inflation forecast for 2021 after UMP announcement.
- Key coefficient patterns and diagnostics (numeric values preserved):
  - Average 10-year past inflation is consistently positive and statistically significant across most specs (e.g., Spec 1: 1.081*** (0.129); Spec 4: 1.303*** (0.160); Spec 7: 1.133*** (0.170)).
  - DGF dummy coefficients are generally negative and not statistically significant (examples: Spec 1: 0.260 (0.480); Spec 7: 0.435 (0.593)).
  - UMP size (in % GDP) coefficients are small and not statistically significant across specs (example Spec 1: 0.0207 (0.0526); Spec 7: 0.0431 (0.0605)).
  - Mauritius dummy large positive and often significant (Spec 1: 3.078* (1.542); Spec 3: 3.112** (1.530); Spec 6: 2.636*** (0.416)).
  - Observations: Spec 1: 64, Spec 2: 64, Spec 3: 64, Spec 4: 47, Spec 5: 48, Spec 6: 47, Spec 7: 51.
  - R-squared: Spec 1: 0.837, Spec 2: 0.838, Spec 3: 0.839, Spec 4: 0.870, Spec 5: 0.863, Spec 6: 0.971, Spec 7: 0.854.
- Interpretation reported in source:
  - The regressions control for prior forecast, average ten-year inflation, fiscal conditions, output gap, and central bank transparency.
  - Results indicate lagged dependent variable and average past inflation are key determinants of 2021 inflation expectations; neither type (DGF) nor size of UMP programs has a significant effect in baseline and most alternative specifications.

*IMF Departmental Paper — Monetary Finance: Do Not Touch, or Handle with Care? (content unit: mfdnthcea)*

### Executive Summary ...............................................................................................iv

### mfdnthcea - Executive Summary

### Context and motivation
- Over the past decade and a half, the world economy has confronted two major crises—the global financial crisis (GFC) and the COVID-19 pandemic.
- In both crises central banks cut interest rates and deployed unconventional monetary policy tools; many countries were pushed into a liquidity trap with interest rates close to zero while public debt rose to historic highs.
- By the end of 2020, policy rates were below 1 percent in more than 60 percent of the global economy.
- The debate centers on whether central banks should reconsider the prohibition on monetary finance (MF) — defined here as the financing of the government via a permanent increase in the monetary base beyond the level consistent with the inflation target and without paying interest on the newly created monetary base.

### Theoretical arguments and distinctions
- Proponents’ arguments:
  - A fiscal stimulus financed with money creation could have a stronger effect on aggregate demand than a debt-financed one because MF does not increase public debt or the associated expected future tax burden.
  - A permanent increase in the monetary base should stimulate inflation and reduce real rates.
  - MF can, in some models, be used to avoid self-fulfilling runs on public debt (relevant for risks of sovereign debt crises).
- Opponents’ arguments:
  - MF risks fiscal dominance and could undermine central bank independence and credibility.
  - If central banks reveal tolerance for MF, fiscal authorities may press for further support beyond macro stabilization needs, potentially leading to higher inflation expectations and run-away inflation.
  - Critics argue MF may simply swap government debt for central bank liabilities and carry limited benefits if reserves are remunerated.
- Distinction emphasized between MF and quantitative easing (QE): conceptual differences are discussed in the paper (Table 1 referenced).

### Empirical strategy to assess inflationary risks
- Two complementary empirical approaches are used given the difficulty of identifying clear historical MF episodes:
  1. A large panel analysis examining the association between money growth (monetary base) and subsequent inflation with data going back to the 1950s.
  2. An examination of whether unconventional monetary policy (UMP) announcements during the COVID-19 pandemic affected inflation expectations, with a focus on emerging market and developing economies (EMDEs).
- Rationale for focusing on EMDEs in the UMP announcement analysis:
  - Several EMDE central banks engaged in operations resembling MF during 2020 (purchases of government bonds in primary markets, provision of government loans/transfers with explicit fiscal support objectives).
  - Central bank independence tends to be less entrenched in EMDEs, making them a better testing ground for MF-induced inflation-expectation effects.

### Key empirical findings
- Panel evidence on money growth and inflation:
  - The strength of the relation between monetary base growth and inflation varies significantly with initial inflation, central bank independence, and fiscal position.
  - When inflation is high, central bank independence is weak, or the fiscal deficit is large (conditions signaling a heightened risk of fiscal dominance), increases in the monetary base are followed by considerable price increases.
  - When inflation is low, central bank independence is strong, and fiscal positions are healthy, increases in the monetary base tend to be followed by modest price increases.
  - Significant non-linearities are detected: inflationary pressures increase more than proportionally with the size of the monetary expansion.
- UMP announcements during COVID-19 and inflation expectations:
  - The paper does not find evidence that UMP announcements in 2020 systematically led to increases in inflation expectations in the EMDE sample studied.
  - Important caveats: these programs were modest in size and launched in response to an exceptional shock, likely perceived as one-off operations.

### Policy implications and recommendations
- The decades-long taboo against MF contributed to establishing central bank independence and a barrier against fiscal dominance.
- Recent theoretical analyses suggest limited circumstances where MF might be beneficial (e.g., persistent liquidity traps or prevention of self-fulfilling sovereign debt crises).
- Given empirical non-linearities and historical examples of destructive outcomes under fiscal dominance, any experimentation with MF should:
  - Remain modest in size.
  - Be limited to countries with low inflation and sustainable fiscal positions.
  - Be used only in exceptional circumstances (persistent liquidity trap or risk of self-fulfilling sovereign debt crisis).
  - Be decided independently by central banks with the sole aim of enhancing macroeconomic stability.
- Historical evidence warns that MF under fiscal dominance can have devastating economic and social consequences; extreme caution is warranted.

*IMF Departmental Paper — Executive Summary*

### 2. Theoretical Underpinnings of Monetary Finance

### 2. Theoretical Underpinnings of Monetary Finance

### Overview
- Chapter describes how MF operates in theory, differentiating MF aimed at stimulating the economy and MF aimed at avoiding self-fulfilling debt crises.
- Discusses: policy objectives, transmission channels, effects on central banks’ balance sheets, possible tensions with inflation targeting, and associated risks.
- Starts by reviewing rationale and effects of QE and summarizes conceptual differences between MF and QE.
- Theoretical discussion assumes presence of nominal rigidities so prices do not adjust instantly to monetary policy shocks.

### A. Quantitative Easing (QE)
- Mechanics and intent:
  - Central bank purchases large quantities of securities, generally long-term government bonds, via open-market operations exchanging bonds for cash in the form of newly created bank reserves.
  - QE involves an increase in the monetary base.
  - QE is used to provide monetary stimulus in a liquidity trap when policy rates have reached the effective lower bound (ELB).
  - Once economy exits liquidity trap and inflation increases back to target, central bank is expected to reduce holdings of government bonds and undo expansion in the monetary base (or start to remunerate reserves).
  - QE does not involve departure from inflation targeting because central bank retains discretion to modulate asset purchases and interest rates to keep inflation at target.
- Transmission channels:
  - Reduces long-term yields through: (1) signaling that policy rates will remain low for extended period; (2) affecting prices when markets are segmented and supply of government bonds is reduced.
- Empirical evidence:
  - Literature suggests QE is effective in reducing government bond yields, especially in periods of scarce liquidity and strong market segmentation.
  - Studies also suggest QE has stimulative effects on output and inflation, though results are less definitive.
- Constraints and risks:
  - Constraint: once yield curve is flat at ELB, long-term bond purchases cannot reduce yields further.
  - Purchasing private sector assets exposes central bank to credit risk.
  - Concerns about excessive inflation proved unfounded in advanced economies after GFC and post-pandemic episodes.
  - QE shortens maturity of consolidated public debt and causes central bank losses when policy rates later increase.
  - Central bank losses generally do not materially affect monetary policy conduct, but large losses could invite political scrutiny or raise fiscal dominance concerns in less-independent central banks.
  - QE can increase private sector leverage and financial stability risks, and may exacerbate inequality via asset price boosts; employment gains can offset inequality effects.

### Conceptual Differences Between Quantitative Easing and Monetary Finance (Table 1 reproduced as statements)
- Quantitative easing
  - Goals: Macroeconomic stimulus
  - Effects on Central Bank Balance Sheets: Large temporary expansion
  - Departure from Inflation Targeting?: No
  - Main Risks: Central bank losses
- Monetary finance
  - Goals: Macroeconomic stimulus; Prevent self-fulfilling crises
  - Effects on Central Bank Balance Sheets: Modest permanent expansion; Potentially large expansion off-equilibrium
  - Departure from Inflation Targeting?: Yes; Only off-equilibrium
  - Main Risks: Fiscal dominance; Central bank losses, fiscal dominance

### B. Monetary Finance for Macroeconomic Stimulus
- Core idea:
  - Central bank generates a permanent increase in the monetary base that can be passed to the public, commonly via fiscal authority actions (tax cuts, temporary spending increases) funded by central bank transfers to the Treasury or by the central bank purchasing and holding government bonds indefinitely.
  - Analogy: Friedman’s “helicopter” money.
- Historical and academic context:
  - Stagflation of the 1970s led to opposition to MF due to association with fiscal dominance.
  - Renewed interest after Japan’s liquidity trap in the 1990s and post-2008 GFC, and during the COVID-19 pandemic.
- Transmission and Ricardian equivalence:
  - MF can overcome Ricardian equivalence because fiat money differs from government bonds: (1) money does not pay interest, reducing government interest bill relative to bond-financed stimulus; (2) money is irredeemable—government never has to increase taxes to retrieve outstanding stock of money.
  - Considerations:
    - Liquidity trap: government bonds pay interest rate close to zero; MF generates interest savings over debt-financed stimulus only because increase in monetary base persists after exit from liquidity trap when interest rates become positive.
    - Remuneration of central bank reserves: MF entails interest rate savings only if central bank reserves—or the additional reserves created through MF—are not remunerated. If reserves are remunerated, MF replaces government bonds with equally remunerated reserves and fails to deliver fiscal savings.
    - Timing of fiscal stimulus: MF is generally associated with immediate fiscal stimulus, but a permanent increase in monetary base can boost aggregate demand even without contemporaneous fiscal stimulus if agents are forward looking.
- General equilibrium perspective:
  - In general equilibrium an increase in monetary base lowers interest rates and raises inflation.
  - In a liquidity trap MF can stimulate demand by lowering real rates through higher expected future inflation.
  - Two insights:
    - MF can be interpreted as an interest rate rule that delivers desired increase in monetary base.
    - Outside a liquidity trap the central bank cannot choose the monetary base and set policy rates as independent instruments; committing to a permanent increase in monetary base implies a temporary departure from an interest-rate rule that keeps inflation at target, accepting temporary inflation above target.
- Comparison with QE:
  - Both aim to provide macro stimulus in liquidity trap and both increase monetary base; key distinction is whether expansion is expected to be unwound (QE) or perceived as permanent (MF).
  - QE: central bank expected to reduce monetary base once economy exits liquidity trap (via bond sales, not rolling over maturing bonds, or by remunerating reserves); thus QE does not entail departure from inflation targeting.
  - MF: increase in monetary base is expected to be permanent; central bank implicitly commits not to tighten when exiting liquidity trap, accepting temporary inflation above target.
- Why consider MF vs QE:
  - MF expected to provide stronger macro stimulus because permanent monetary base increase shapes future real interest rates; QE compresses rates during liquidity trap but not afterward if unwound.
  - Stronger impact of MF is especially relevant when yield curve is flat and QE has limited scope to reduce term premia.
  - MF can deliver desired stimulus with a smaller but more prolonged expansion of monetary base and central bank balance sheet, reducing eventual capital losses and footprint in asset markets.
  - Allowing inflation to rise above target via MF lowers real interest rates and alleviates debt burdens—appealing for countries with high public debt and private sector leverage.
- Commitment and credibility issues:
  - Distinction between QE and MF is largely about announced policy intentions and expectations rather than specific implementation tools.
  - MF may fail if central bank cannot credibly commit to permanent increase in monetary base; if agents expect reversal, MF effects resemble QE.
  - Opposite risk: MF may trigger runaway inflation and fiscal dominance if governments pressure for continued MF, de-anchoring inflation expectations and potentially increasing real rates demanded by investors.
  - Risks increase with larger fiscal pressures and weaker central bank independence; governance and transparency of financing modalities (primary vs secondary market purchases, loans, grants, etc.) matter for perceived fiscal dominance.

### C. Monetary Finance to Prevent Self-Fulfilling Debt Crises
- Rationale and mechanics:
  - MF can be used as sovereign backstop to prevent coordination failures and self-fulfilling debt crises (example: ECB’s 2012 “whatever it takes” episode).
  - If markets coordinate on bad equilibrium, high sovereign yields worsen fiscal position and tighten local financial conditions, increasing default probability; credible central bank commitment to purchase distressed sovereign bonds by expanding monetary base can shift markets to good equilibrium with lower yields and sustainable debt.
- Literature insights:
  - MF backstops can operate as off-equilibrium path: credible commitment prevents runs; central bank may need to show resolve via some purchases or direct financing.
  - Inflation tolerance: backstop credible only if central bank has some tolerance for inflation; must be ready to endure sufficient inflation to prevent default if markets coordinate on bad equilibrium.
  - Sterilized backstops: central bank can provide support via sterilized operations (non-inflationary) such as purchasing bonds using foreign exchange reserves, selling other countries’ bonds in a monetary union, or issuing interest-bearing liabilities (remunerated reserves) to finance bond purchases—these options assume investors do not fear de facto central bank default via financial repression.
- Risks:
  - Central bank losses: risk of shifting default losses from private investors to the central bank if crisis reflects fundamentals rather than a self-fulfilling run.
  - Fiscal dominance: once central bank commits to backstops, government pressure could force continuous MF even when spreads reflect fundamental concerns, compromising inflation control.
  - Required inflation to avert default may be large; example model evidence finds inflation may have to increase to double digits for a protracted period to fend off default risks.

*Source: IMF Departmental Paper — 2. Theoretical Underpinnings of Monetary Finance*

### 3. Historical Evidence on the Association

### 3. Historical Evidence on the Association Between Money and Inflation

### A. Empirical approach
- Analysis assesses association between monetary aggregates and inflation using flexible local projections à la Jordà (2005).
- Regression links money growth at time t in country i, d logMit, to inflation h years ahead, πit+h, controlling for:
  - Lagged values of money growth and inflation (up to l = 10 lags).
  - Growth rate of real GDP, d logYit (up to l = 10 lags).
  - Country and year fixed effects κi + τt.
- Specification estimated on annual data from the 1950s to 2020 for a panel of up to 157 countries.
- Two monetary aggregates considered:
  - Monetary base (MB) — currency in circulation and bank reserves at the central bank.
  - Money supply measured by M2 — currency in circulation and deposits in checking and saving accounts.
- Key identification concerns:
  - MB changes are set by monetary authorities and hence endogenous to economic conditions.
  - M2 fluctuations reflect both policy (money supply) and behavior (money demand).
  - Controls for recent money growth, inflation, and real GDP growth are used to mitigate endogeneity, but concerns remain.

### B. Empirical findings on monetary expansions and inflation
- Impact of a 10 percent monetary increase:
  - A 10 percent increase in MB is associated with an increase in inflation by about 1.5 percentage points on impact and in the subsequent year.
  - A 10 percent increase in M2 is associated with an increase in inflation by about 2 percentage points (point estimates).
  - After 10 years, a 10 percent increase in MB and M2 is associated with an increase in the price level by about 6 and 8 percent, respectively.
  - The 90 percent confidence bands for M2 are much wider than for MB.
- Nonlinearity by size of monetary expansion:
  - A 1 percent increase in MB is associated with a 0.3 percent increase in the price level after 10 years.
  - A 100 percent increase in MB is associated with an increase in the price level up to 94 percent after 10 years.
  - Implication: inflationary risks may increase more than proportionally with the size of the monetary expansion.
- Dependence on initial inflation:
  - Interaction of money growth with one-year lagged inflation is positive and statistically significant on impact.
  - Examples:
    - If initial inflation is 100 percent, a 10 percent increase in MB is associated with an additional increase in inflation by 9 percentage points.
    - If initial inflation is 2 percent, a 10 percent increase in MB is associated with at most a 0.9 percentage point increase in inflation one year ahead.
  - Interpretation: higher initial inflation amplifies the inflationary response to monetary expansions.
- Role of central bank independence:
  - Interaction between money growth and the de jure index of central bank independence (Garriga 2016) is negative and statistically significant.
  - A low level of independence (25th percentile) implies an increase in inflation three times as large as for a high independence level (75th percentile) after a given MB increase.
  - Implication: a highly independent central bank can sustain an increase in the monetary base three times as large as a low-independence central bank while experiencing the same price pressures.
- Role of fiscal balance:
  - Interaction of money growth with a dummy for fiscal deficit > 3 percent of GDP is positive and significant.
  - When fiscal deficit is below 3 percent there is no statistically significant association between MB and inflation.
  - When fiscal deficit exceeds 3 percent—with an average of 6.2 percent in the regression sample—an increase in MB is associated with a significant and persistent increase in inflation.
- Robustness checks:
  - Results robust to dropping observations with annual inflation above 100 percent.
  - Results unchanged when removing quantitative easing episodes in advanced economies.
  - Interactions of money growth with initial inflation, central bank credibility, and fiscal deficit jointly included yield similar results.
  - Similar qualitative influences observed for M2, though estimates are less precise.

### C. Implications for Monetary Finance (MF) policy
- Inflationary risks from MF are larger for larger increases in the monetary base; recommended approach:
  - Central banks considering MF should experiment starting from a modest scale.
- Contexts where inflation risks appear contained:
  - Low initial inflation.
  - Strong central bank credibility/independence.
  - Modest fiscal deficits (below 3 percent of GDP).
- Contexts of elevated risk where MF is likely to trigger sharp price responses:
  - High initial inflation.
  - Weak central bank independence.
  - Large fiscal deficits (above 3 percent of GDP).
- Policy takeaway:
  - Central banks with weak credibility should refrain from embarking on monetary expansions under MF.

### D. UMP announcements during COVID-19 and inflation expectations (overview)
- Alternative assessment examines whether unconventional monetary policy (UMP) announcements during March 2020–December 2020 affected inflation expectations.
- Sample and classification:
  - Analysis covers UMPs in 49 advanced economies and EMDEs between March 2020 and December 2020.
  - Sample includes 15 cases of direct government financing (DGF) and 64 other UMP cases.
  - EMDE programs generally smaller than advanced economies’ programs.
- Notable example cited:
  - Bank of Ghana announced purchase of a Government of Ghana COVID-19 relief bond with a face value of GH¢5.5 billion at the Monetary Policy Rate with a 10-year tenor and a moratorium of two years (principal and interest); Bank of Ghana stated readiness to continue the Asset Purchase Programme up to GH¢10 billion.
- Inflation expectations data and coverage:
  - Real-time inflation expectations from Consensus Economics (monthly; covering 31 countries in the sample).
  - WEO forecast vintages (quarterly; covering the other 18 countries).
  - Focus on end-of-period inflation forecasts.
- Empirical observations on expectations:
  - Average inflation forecasts for 2020 before and after central bank announcements are stable for both DGF and other UMP programs.
  - No notable difference in inflation expectations between relatively small and relatively large UMP programs.
  - Cross-section and panel regressions (reported in Annex 7 and Annex 8) formalize these patterns.

*Source: mfdnthcea - 3. Historical Evidence on the Association, IMF Departmental Papers.*

### Annex 6, contains EMDEs of a wide variety of institutional backgrounds and development levels. For example, among the EM

### mfdnthcea - Annex 6, contains EMDEs of a wide variety of institutional backgrounds and development levels. For example, among the EM

### Sample heterogeneity and key descriptive statistics
- Structural fiscal deficit in the sample ranges from –8.9 to 2.2 percent of GDP.
- Average 10-year past inflation in the sample ranges from 0.8 to 21.4 percent.
- Median program size referenced in analysis: 2% of GDP.
- Bank of Mauritius one-off transfer: equal to almost 15 percent of GDP; end-of-year inflation in 2020 closed at 2.7 percent.

### Effects of UMP (unconventional monetary policy) announcements on inflation expectations
- Baseline explanatory variables include:
  - Latest inflation forecast before the announcement.
  - Country’s average inflation over the past ten years.
  - Size of the UMP program.
  - Dummy for DGF programs.
  - Real-time forecast of the 2020 fiscal deficit at the time of the announcement.
  - Dummy for Mauritius (outlier regarding UMP size).
- Regression results:
  - Lagged dependent variable and average past inflation are key determinants of 2020 inflation expectations (persistence and influence of past inflation).
  - Neither the type (DGF or not) nor the size of UMP programs has a significant effect on the inflation forecast after the UMP announcement in the baseline and most specifications.
  - Specifications 2–7 (which add interactions, alternative fiscal measures, output gap, and central bank transparency) do not find UMP-related variables statistically significant.
- Robustness checks:
  - Table 2 (2021 inflation expectations collected in first survey after announcements): no statistically significant effect associated with UMP and DGF programs across seven specifications.
  - Table 3 (2020 inflation expectations in second survey after announcement): baseline results remain similar; potential delayed effects not evident in baseline.
  - Specifications 4 and 5 indicate an impact of UMP (but not DGF) on inflation expectations in smaller samples; effect not robust in reduced-sample regressions (specifications 4–7).

### Sterilization, monetary base changes, and sterilization robustness tests
- Observed relation: announced UMPs translated nearly one-for-one into changes of the monetary base on average, though this varies across countries.
- Some countries partially sterilized UMPs (e.g., issuance of central bank bills), which partially unwound impacts on the monetary base.
- Robustness tests in Table 4 of Annex 7:
  - Specification 2 adds change in the monetary base divided by total announced UMP size and its interaction with DGF dummy to test sterilization effects.
  - Specification 3 trims sample to UMP cases where the monetary base increased by at least half of the announced UMP size (little sterilization).
  - Neither specification changes the relation between UMPs and inflation expectations relative to baseline.

### Panel analysis and time-series evidence
- Panel data (monthly Consensus Economics inflation expectations) examines 12-month ahead inflation forecast (weighted average of 2020 and 2021 forecasts).
- Baseline panel includes country and monthly time fixed effects.
  - April 2020 dummy is particularly large, significant, and negative (global nosedive in inflation expectations due to COVID-19).
  - Lagged dependent variable is statistically significant; UMP-related variables are not.
- Alternative specifications:
  - Without country fixed effects: resembles cross-section with time-series component; results consistent with cross-section findings.
  - Dissecting UMP/DGF by country type and adding quadratic UMP terms: an interaction between EM dummy and UMP size is statistically significant in one specification, but economically small—"a UMP announcement equal to 1 percent of GDP raises inflation expectations by 0.04 percentage points".
  - Effect concentrated on quantitative easing programs; DGF dummy among EMs not statistically significant (and has a negative coefficient).

### Historical cautionary cases (Box 1)
- Zimbabwe, 2007–08:
  - Peak month-over-month inflation rate reached 7.96 x 10^10 percent.
  - Prices doubled every 24.7 hours at that peak.
  - Budget deficits in 2005–08 were in the range of 25–45 percent of GDP.
  - Outcomes included issuance of a 100 trillion Zimbabwe-dollar note, per capita GDP falling to less than half its level a decade earlier, and 70 percent of the population underfed.
- Comparative note: German hyperinflation of 1922–23 peaked at 29,500 percent (monthly); Hungarian hyperinflation of 1945–46 peaked at 4.19 x 10^16 percent (monthly).
- Suriname, 2015–2020:
  - Central bank extended loans to the government amounting to 11.6 percent of GDP in 2015; inflation rose from 3.9 percent in 2014 to 25.1 percent in 2015 and 52.4 percent in 2016.
  - Fiscal deficit grew from 6 percent of GDP in 2018 to 21.2 percent in 2019.
  - Central bank provided loans worth 15.9 percent of GDP to the government until July 2020.
  - IMF WEO forecasted 108.1 percent end-of-period inflation for 2020; actual 2020 outturn moderated to 61 percent.
  - Real GDP contracted 13.5 percent in 2020.

### Main conclusions and policy recommendations
- Empirical summary:
  - UMP announcements during COVID-19, including those involving direct government financing in EMDEs, did not lead to an economically and statistically significant increase in inflation expectations in the sample and specifications examined.
  - The lack of systemic effect may reflect that these UMPs were modest, often one-off, and implemented in response to a clear, exceptional shock (the pandemic).
  - Cross-country, long-run evidence indicates that increases in the monetary base are associated with modest price pressures only in countries with low inflation, strong central bank independence, and modest fiscal deficits; by contrast, high inflation, weak central bank independence, and large fiscal deficits amplify price responses and risk fiscal dominance.
  - Inflationary pressures tend to increase more than proportionally with the size of the monetary expansion.
- Policy guidance:
  - Further analysis of conditions under which MF could be warranted is merited, but relaxing the taboo carries considerable risks.
  - Any experimentation with MF should be:
    - Limited to extreme circumstances.
    - Modest in size, given documented non-linearities.
    - Considered only by countries with low inflation and sustainable fiscal positions.
  - Crucially, central banks must retain full independence to decide if and when MF might be appropriate to achieve macroeconomic stabilization objectives.
  - In contexts of elevated inflation, unsustainable fiscal positions, and weak central bank independence (signals of heightened fiscal dominance), policymakers should vigorously refrain from using MF because it is much more likely to trigger strong inflationary pressures with disastrous economic outcomes.

*Source: IMF DEPARTMENTAL PAPERS • Monetary Finance: Do Not Touch, or Handle with Care? (content unit: mfdnthcea - Annex 6).*

### Annex 1. Monetary Finance in

### Annex 1. Monetary Finance in Historical Perspective

### The Great Depression
- Pre-Depression view: period of strong economic growth in the United States during the 1920s—soon after the creation of the Federal Reserve System—fostered the idea that advances in monetary technology had unleashed a new era of macroeconomic stability (Friedman 1968).
- Crisis: The severity of the Great Depression shattered this notion; monetary policy seemed unable to provide sufficient macroeconomic support while the economy was mired in a prolonged liquidity trap.
- Keynesian view (Keynes 1936): monetary policy worked merely as a string—monetary tightening could raise interest rates and curb economic growth, but monetary easing was largely ineffective in supporting aggregate demand in a liquidity trap.
- Monetarist challenge (Friedman and Schwartz 1963): argued that the recession’s severity was largely due to an insufficient money supply, evidenced by a prolonged slump in the M2 monetary aggregate; a stronger increase in the monetary base would have offset deflationary pressures.
- Helicopter-money analogy (Friedman 1969): example of “a helicopter” dropping an additional $1,000 in bills from the sky to illustrate money-financed fiscal stimulus; interpreted as a tax cut financed with money creation or bond issuance purchased with newly printed money and rolled over indefinitely.
- Implication: MF (money-financed tax cut or permanent increase in the monetary base) would induce some spending and thus could provide stimulus even in a liquidity trap.

### The Post-WWII Years and the Great Moderation
- Shift in focus: concerns about liquidity traps faded as nominal interest rates rose above zero after World War II; central bankers and academics focused on combating inflationary pressures of the 1970s.
- Time-inconsistency and inflation bias: Kydland and Prescott (1977) and Barro and Gordon (1983) showed central banks have an incentive to create surprise inflation to stimulate growth and erode real public debt, generating a bias toward excessive money creation and inflation.
- Policy response: move toward stronger central bank independence and adoption of inflation targeting regimes to commit credibly to price stability and avoid using MF to lower unemployment or monetize public debt.
- Outcome: monetary dominance with credible commitment to inflation targets delivered low and steady inflation and sustained economic growth until the 2008 GFC, termed the Great Moderation (Blanchard and Simon 2001, Stock and Watson 2003).

### Japan’s Liquidity Trap and Renewed Debate on MF
- Context: Collapse of asset-price bubble in 1991 led to a severe recession and liquidity trap in Japan; large central bank asset purchases and fiscal stimulus produced feeble activity and deflationary pressures.
- Krugman (1998) argument: liquidity trap caused by an “inverse” credibility problem—central banks should “credibly promise to be irresponsible” by committing to a permanent increase in the money supply to allow temporary higher inflation after exiting the trap; expectation of higher future inflation lowers real interest rates and stimulates current demand.
- Formalization: Eggertsson and Woodford (2003) show via a dynamic general equilibrium model that to provide stimulus in a liquidity trap, central banks should commit to a lower interest rate path (i.e., higher money supply) after exiting the trap.
- Reversal problem: Auerbach and Obstfeld (2005) argued asset purchases by the Bank of Japan were likely undermined by expectations these operations would be reversed.
- Fiscal–monetary coordination: Eggertsson and Woodford (2003), Auerbach and Obstfeld (2005), and Eggertsson (2006) argue matching money supply increases with fiscal stimulus reinforces expectations that monetary expansion is permanent because fiscal expansion raises public debt and incentives for future monetization.
- Bernanke (2002, 2003): called for tight cooperation between monetary and fiscal authorities; fiscal policy alone may be weak if public debt is elevated and Ricardian equivalence reduces stimulus; financing deficits with a permanent increase in the monetary base would strengthen stimulus by removing expectation of higher future taxes.

### Key Analytical Insights (historical)
- MF can overcome Ricardian equivalence in certain environments by altering the PDV of taxes through:
  - Interest-rate savings: if money pays a lower interest rate than government bonds, money creation reduces government interest bills relative to issuing bonds, lowering the PDV of taxes.
  - Irredeemable property: if lim_{j → ∞} Λ_{j,t} M_{j+1} > 0 (i.e., the growth rate of money exceeds the nominal discount factor), a permanent increase in the money stock allows for lower taxation because money is an irredeemable liability the government never repurchases.
- Price-stickiness channel: degree of price stickiness determines whether higher nominal household spending from MF translates into higher prices (if prices fully flexible, output unchanged) or temporarily higher output (if prices slow to adjust).

*Source: Annex 1. Monetary Finance in Historical Perspective, from the IMF Departmental Paper "Monetary Finance: Do Not Touch, or Handle with Care?"*

### Annex Table 7.2. Dependent Variable: First Inflation Forecast for 2021 After UMP Announcement

### Annex Table 7.2. Dependent Variable: First Inflation Forecast for 2021 After UMP Announcement

### Regression specifications and coefficients (Specs 1–7)
- Last inflation forecast before announcement
  - Spec 1: 0.0982 (0.199)
  - Spec 2: 0.106 (0.202)
  - Spec 3: 0.0596 (0.203)
  - Spec 4: 0.328 (0.259)
  - Spec 5: 0.337 (0.272)
  - Spec 6: 0.967*** (0.0894)
  - Spec 7: 0.225 (0.260)
- Average 10-year past inflation
  - Spec 1: 1.081*** (0.129)
  - Spec 2: 1.084*** (0.131)
  - Spec 3: 1.046*** (0.135)
  - Spec 4: 1.303*** (0.160)
  - Spec 5: 1.263*** (0.166)
  - Spec 6: 0.0673 (0.0696)
  - Spec 7: 1.133*** (0.170)
- DGF dummy
  - Spec 1: 0.260 (0.480)
  - Spec 2: 0.0980 (0.760)
  - Spec 3: 0.243 (0.478)
  - Spec 4: 0.296 (0.614)
  - Spec 5: 0.739 (0.609)
  - Spec 6: 0.0560 (0.155)
  - Spec 7: 0.435 (0.593)
- UMP size (in % GDP)
  - Spec 1: 0.0207 (0.0526)
  - Spec 2: 0.0237 (0.0541)
  - Spec 3: 0.0018 (0.0547)
  - Spec 4: 0.0652 (0.0553)
  - Spec 5: 0.0589 (0.0571)
  - Spec 6: 0.00196 (0.0141)
  - Spec 7: 0.0431 (0.0605)
- DGF * UMP size
  - Spec 1–5,6: not reported
  - Spec 7: 0.0499 (0.180)
- Fiscal deficit
  - Spec 1: 0.0184 (0.0633)
  - Spec 2: 0.0133 (0.0664)
  - Spec 3: 0.00842 (0.0171)
  - Spec 4: 0.00502 (0.0714)
  - Specs 5–7: not reported
- Mauritius dummy
  - Spec 1: 3.078* (1.542)
  - Spec 2: 3.541 (2.280)
  - Spec 3: 3.112** (1.530)
  - Spec 4: 2.866* (1.594)
  - Spec 5: 3.369** (1.635)
  - Spec 6: 2.636*** (0.416)
  - Spec 7: 2.789 (1.660)
- Gross government debt
  - Reported only in Spec 6: 0.00501 (0.00576)
- Structural deficit
  - Reported only in Spec 6: 0.0361 (0.0760)
- Cyclically adjusted deficit
  - Reported only in Spec 6: 0.0723 (0.0774)
- Output gap
  - Reported only in Spec 6: 0.0528** (0.0259)
- Central bank transparency
  - Reported only in Spec 6: 0.105 (0.0953)
- Constant
  - Spec 1: 0.0834 (0.494)
  - Spec 2: 0.113 (0.509)
  - Spec 3: 0.458 (0.509)
  - Spec 4: 0.331 (0.521)
  - Spec 5: 0.0318 (0.529)
  - Spec 6: 0.477** (0.206)
  - Spec 7: 0.903 (1.126)

### Key statistics and diagnostics
- Observations by specification: Spec 1: 64, Spec 2: 64, Spec 3: 64, Spec 4: 47, Spec 5: 48, Spec 6: 47, Spec 7: 51
- R-squared by specification: Spec 1: 0.837, Spec 2: 0.838, Spec 3: 0.839, Spec 4: 0.870, Spec 5: 0.863, Spec 6: 0.971, Spec 7: 0.854
- Standard errors in parentheses; significance: ***p  0.01, **p  0.05, *p  0.1
- Note: fiscal and output gap variables are WEO forecasts.

### Findings and interpretation (as stated in the source)
- The regression table analyzes whether the size and type of a UMP announcement had an impact on the first inflation forecast for 2021 after the announcement.
- The regressions control for factors that can influence the inflation forecast, including:
  - the inflation forecast prior to the announcement,
  - the average inflation rate in the prior ten years,
  - fiscal conditions,
  - the output gap,
  - central bank transparency.

*IMF DEPARTMENTAL PAPERS • Monetary Finance: Do Not Touch, or Handle with Care? — Annex Table 7.2.*

---


_Source: https://www.imf.org/-/media/files/publications/dp/2022/english/mfdnthcea.pdf_