## wpiea2021006-print-pdf

## Source details

**Canonical URL:** [wpiea2021006-print-pdf](https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021006-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2021/english/wpiea2021006-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2021/english/wpiea2021006-print-pdf.pdf.json)

---

### Introduction: context, contention, and agenda
- Massive expansion of central bank balance sheets since 2009 in countries with well-developed financial markets has not yet led to accelerating inflation.
- Resurgence of central bank balance sheet growth and increase in monetary aggregates associated with policy responses to the COVID-19 crisis has reignited debate over whether current central bank actions will cause a spike in inflation if not swiftly reversed.
- Two diametrically opposed views:
  - Some argue central bank balance sheet expansion should be curtailed to short-circuit money growth and inflation.
  - Others, including some associated with “modern monetary theory”, advocate increased reliance on monetary (vs debt) finance during the COVID crisis and for future large-scale public investment, arguing there remains ample scope to employ monetary finance without inflationary consequences.
- Core theoretical contention:
  - Whether the quantity of money is exogenous (determined by the state: central bank and treasury) or endogenous (determined by private sector demand and supplied passively by the state).
  - The state directly controls only the forms of money it issues: currency and deposits held by the private sector at the central bank (central bank or outside money, M0, high powered money, monetary base).
  - Key empirical operational question: does the state determine the amount of private sector deposits held at the central bank—usually synonymous with bank reserves?
- Historical operational change:
  - Prior to the Global Financial Crisis (GFC), central banks generally provided the quantity of reserves demanded by banks, used interest rates rather than quantitative policy instruments, and did not target monetary aggregates.
  - Since 2009, in floating exchange rate regimes central bank asset purchases have determined the quantity of the monetary base (making the monetary base “exogenous”).
  - Once central banks reached the effective lower bound on nominal interest rates, they purchased assets in unconventional policy actions and allowed corresponding excess bank reserves to remain in the system.
- Motivations for revisiting money–inflation relationship:
  - Beliefs matter for inflation expectations and debt costs.
  - Regime change in money supply determination since the GFC may have altered empirical relations.
  - Policy relevance regarding monetary finance: distinguishing between endogenous money growth and exogenous monetary finance and when monetary finance is inflationary.
- Empirical and policy agenda:
  - Review modern history of monetarism and alternatives.
  - Show changing empirical relationship of various measures of money and inflation; earlier naïve correlations have largely disappeared.
  - Explain why disappearance of correlations does not validate permanent monetary finance.
  - Anticipate structural forces: rapid technological innovation in payments systems and global pledged collateral markets portend a declining demand for central bank liabilities.

### I. Quantity Theory of Money—theory, equation of exchange, and payments technology
- Four interrelated QTM propositions highlighted:
  - (i) proportionality of money and prices,
  - (ii) exogeneity of nominal money stock and its causal role,
  - (iii) monetary theory of the price level,
  - (iv) neutrality of money.
- Equation of Exchange formulations and interpretations:
  - Fisher (1922): MV(t) = P(t)T
  - Monetarist transformation: MV(y)=P(y)Y
  - Cambridge k formulation: M = k(y)P(y)Y
  - Velocity (V) defined as turnover rate of money; 1/V interpreted as k in Cambridge approach.
- Conceptual problems with income-based velocity:
  - Money is used across many markets with different payment systems and velocities; income velocity is not a meaningful measure of overall money turnover.
  - Examples:
    - The ratio of national income in the United States to currency: about 18.
    - The ratio of U.S. interbank payments to bank reserves: 77,874 in 2007.
- Payments technology and empirical consequences:
  - Wholesale payments systems experienced large speed and efficiency gains earlier; retail payments also accelerated since the 1980s.
  - Trading in financial markets has grown much faster than the real economy since the 1950s.
  - Wholesale payments (Fedwire, CHIPS, CHAPS, TARGET2) are approximately 50X payments made for final goods and services in the U.S., UK, and Eurozone.
  - Payments for final goods and services comprise less than one percent of total payments.
  - Interpreting a truly transactions-based price index would be extraordinarily difficult.

### Key empirical statistics on payments, velocities, and monetary composition
- Exact figures preserved from source:
  - The ratio of national income in the United States to currency: about 18.
  - The ratio of U.S. interbank payments to bank reserves: 77,874 in 2007.
  - Payments in the main interbank wholesale markets in the U.S., UK, and Eurozone (Fedwire, CHIPS, CHAPS, and TARGET2) are approximately 50X payments made for final goods and services.
  - The sum of payments effected in 2017 through Fedwire and CHIPS: 1 quadrillion 133 trillion dollars; U.S. GDP that same year: US$19.5 trillion.
  - Conventional income velocity of M1 (GDP divided by M1) was between 3 and 5 from 1927 to 1977.
  - U.S. DD (demand deposit) turnover ratio history:
    - 41 prior to the Great Depression,
    - rises sharply thereafter,
    - 922 in 1996 (the year the Federal Reserve ceased publishing the data).
  - A turnover ratio of 922 on an annual basis implies, given approximately 250 business days per year, that the average deposit balance turns over almost 4 times per day.
  - Total payments made with DD were 58x U.S. GDP in 1996.
  - If one assumes that 85 percent of U.S. GDP was then purchased with debits of U.S. DD (assumption presented contextually in source).
  - 1.5 percent of all payments made with DD were then made for the acquisition of final goods and services.
  - Income velocity of currency (C) in the United States has not been rising during the last half of the 20th century, leading to a rising proportion of C in M1.
  - C exceeded DD in M1 in the U.S. in 1997 and remained so for twenty years—C was more than double DD just before the GFC in 2007-08.
- Wholesale payments, reserves, and extreme turnover differences (preserved numbers):
  - 1947: US$ 316 billion in payments through Fedwire; stock of reserves at the FRBs US$ 17.9 billion — yielding an annual turnover of 58.
  - 2007: US$ 670.7 trillion in payments through Fedwire; reserves amounted to US$ 9.0 billion — yielding an annual turnover of 77,784.
  - Payments increased by 212,115 percent while reserves fell by 50 percent (2007 vs 1947 comparison as presented).
  - By 2007, C represented about 98 percent of M0 although estimated at most involved in US$ 2 trillion in payments that year while payments made through the two U.S. LVPS, CHIPS and Fedwire, during the same year amounted to US$ 1 quadrillion 156 trillion.
  - 2007 average stock of U.S. currency outstanding: US$ 811.7 billion; assuming 50 percent is held outside the United States gives an annual turnover ratio of currency of about 5.
  - Average 2007 stock of reserves: US$ 8.6 billion giving a gross LVPS turnover ratio of 134,453 per year.

### Heterogeneous velocities, endogeneity of reserves, and operational practice
- Various subcomponents of M0 and M1 have wildly different velocities and stability properties: between C and reserves (M0), and C and DD (M1).
- M0 remains the key monetarist variable as notionally controlled by the central bank and thus exogenous; alternative labels: base money, high-powered money, outside money.
- Currency is widely not rationed by central banks; almost no central bank prevents recipients from returning currency to the banking system.
- Reserves differ from currency: only banks may hold reserves; reserves are assets, not monetary liabilities of banks, and are included only in M0.
- Textbook money multiplier (1/rr) depends on reserve requirements and the ratio C/DD; in reality these relationships have varied considerably over time.
- Prior to the GFC advanced-country central banks provided as much currency and reserves as desired at the opportunity cost of a target overnight interest rate; they set the price, not the quantity, of reserves.
- Daylight overdrafts and RTGS operation imply reserves increase during the operating day and fall at the end of the day — the reserves multiplier is unstable and highly variable intraday.
- Example intraday/firm evidence:
  - Reported total U.S. funds transfer daylight overdrafts of US$ 76 billion per day compared with total reserve balances around US$ 35 billion (Mengle, Humphrey, Summers (1987) cited).
  - 2006 JP Morgan Chase held US$ 1,352 billion in assets of which only US$ 2.2 billion (0.16 percent) were reserves at FRBs.
  - Fed’s decision in March 2020 to set the required reserve ratio at zero effectively eliminated reserve requirements for all depository institutions.

### Empirical breakdowns of money–price proportionality (selected coefficient results)
- Re-estimation of McCandless and Weber methodology shows departures from proportionality between money growth and inflation, especially for reserves.
- Table 2 (1960–1989): Full Sample (88 countries) coefficients:
  - M0 0.9461
  - M1 0.9398
  - M2 0.9299
  - Currency (C) 0.9604
  - Reserves (M0-C) 0.6436* (* indicates statistically different from 1 at 5 percent)
  - D.Deposits (M1-C) 0.8905*
  - Other Deposits (M2-M1) 0.9258
- Table 3 (1990–2019): Full Sample (143 countries) coefficients:
  - M0 0.7504*
  - M1 0.7728*
  - M2 0.7787*
  - Currency (C) 0.7841*
  - Reserves (M0-C) 0.5484*
  - D.Deposits (M1-C) 0.7250*
  - Other Deposits (M2-M1) 0.7342*
  - OECD (20) shows negative coefficient on reserves (-0.3130*).
- Table 4 (2001–2019): Full Sample (140 countries) coefficients:
  - M0 0.6685*
  - M1 0.72768*
  - M2 0.8225*
  - Currency (C) 0.6328*
  - Reserves (M0-C) 0.4006*
  - D.Deposits (M1-C) 0.6573*
  - Other Deposits (M2-M1) 0.7606*
  - OECD (18) shows reserves -0.3214*.
- Key empirical finding: the coefficient on Reserves (the variable most plausibly exogenous under monetarist logic) is the lowest of all components and often statistically different from 1, including negative in some OECD samples — contradicting a simple causal link from reserves to inflation.

### Country evidence and interpretation
- Japan QE (2001–2006) targeted current account deposits (reserves) as the operational target.
  - Correlations and figures for Japan (1990–2019) show essentially zero correlation between M0 and inflation.
  - Change in bank reserves lagged by 12 months and bank credit expansion during 1991-2020 is exactly zero (as presented).
  - Conclusion: expansion in reserves had neither a commensurate impact on credit nor directly on inflation in Japan.
- Post-GFC reserve expansion:
  - Superabundant reserves in the United States, UK, Eurozone, Switzerland and Japan have not been associated with marked expansions in credit nor significant inflation increases.
  - After the GFC, capital requirements increased and leverage ratios were imposed/raised; capital availability became a more important constraint to lending than reserves.
  - In some cases increases in reserves combined with prudential ratios may have paradoxically constrained bank lending.
  - Some jurisdictions exempted central bank deposits from leverage ratio calculations (UK), others relaxed ratios in response to COVID-19 (United States).

### Theoretical distinction: asset swaps vs permanent monetary finance
- Open-market purchases swap treasury securities for reserves; Wallace (1981) shows such swaps are theoretically irrelevant for the economy — reserves versus short-term treasury claims are close substitutes.
- Monetarist “helicopter money” involves money entering the economy without an offsetting asset swap (permanent money finance tied to fiscal transfers), which would plausibly have a stronger inflationary effect.
- Distinction central for policy:
  - CBDL created in exchange for bonds (asset-swap QE) tends to change asset composition without causing the inflationary outcomes associated with permanent monetary finance.
  - CBDL created to permanently finance fiscal spending is the form most likely to be inflationary.

### Implications for monetary finance and policy takeaways
- Recent large increases in CBDL (reserves and currency) have not caused inflation because much of the increase substituted for financial assets rather than representing permanent, unrequited transfers to the public.
- Advocates of permanent monetary finance who infer non-inflationary outcomes from recent CBDL increases conflate two ways money can enter the economy.
- Unsustainable fiscal expenditures will be inflationary regardless of whether financed initially with money or bonds.
- Excess reserves beyond payment-system needs are “deadweight” on banks’ balance sheets: reserves can only be held by banks, are poor collateral, and are subject to capital requirements; treasury securities are more efficient and tradeable widely.
- Historical development of domestic debt markets in many countries makes debt issuance a preferred way to finance deficits relative to returning to broad monetary finance.
- Three fundamental points emphasized:
  - (i) The relationship between the quantity of CBDL and inflation has completely broken down and shows no causal relationship in modern data.
  - (ii) Recent exogenous increases in CBDL (policy-driven) have not altered (i) and have made the lack of causal linkage clearer.
  - (iii) Permanent monetary finance should not be the preferred route in the near-zero (or negative) rate environment.
- Monetary aggregates are tangential to central banks’ objectives in practice; interest-rate policy remains the operational focus for most central banks.
- Given secular declines in demand for CBDL and technological evolution of payments, policy relying on increased permanent monetary finance runs counter to historical and technological trends.

### Annex 1 — Collateral and Money Aggregates: role, IS-LM enrichment, and policy trade-offs
- Collateral is central to money markets and market plumbing (repo, securities lending, prime brokerage financing, derivatives margins).
- QE removes good collateral—typically sovereign bonds, but even corporates and equities—from the market, with implications for associated rates and market functioning.
- Enriched LM specification proposed: M = f(Y, r, c_p), where c_p measures the efficiency of pledged collateral (that can be reused).
  - c_p governs the efficiency of monetary policy transmission given collateral’s lubricant role in secured finance markets.
  - When collateral markets are constrained (c_p declines), for a given M a larger change in r is required to produce the same change in Y.
  - A shrinkage in collateral supply reduces the efficiency of monetary policy transmission.
- Dynamics under collateral constraints:
  - The LM curve can pivot and intersect IS at different points depending on slope and c_p; QE may increase M while lowering c_p by draining good collateral—producing countervailing effects on output and rates.
- Empirical/market examples and numbers:
  - After the Lehman crisis, pledged collateral volumes (including price effects) fell from around US$10 trillion before the crisis to around US$5.5 trillion afterwards; volumes are inching back slowly to US$8.5 trillion.
  - Market dysfunction episodes: dealers shied away from making a market in U.S. Treasury repo in September 2019 and avoided making a market in off-the-run U.S. Treasuries in March 2020, highlighting dealer balance sheet constraints that limit pledged collateral reuse rate even with large debt issuance.
- Transmission and policy implications:
  - When collateral use drops, financial intermediation slows; investors concerned about counterparty risk may lend fewer securities, leaving idle collateral in segregated accounts.
  - A widening of eligible asset pools (increasing pledgeability) or large-scale QE (draining good collateral) both alter c_p and market functioning.
  - Policymakers face a trade-off between constraining collateral markets (lower c_p, impaired transmission, market plumbing stress) and positive output effects of QE (higher M).
  - Rebound in pledged collateral reuse rates depends on dealer balance sheet space; large debt issuance alone may not restore market plumbing without addressing balance-sheet capacity.

*Source: wpiea2021006-print-pdf — IMF PDF chapter content provided.*

### Introduction ...........................................................................................................

### Introduction

### Context and central question
- Massive expansion of central bank balance sheets since 2009 in countries with well-developed financial markets has not yet led to accelerating inflation.
- Resurgence of central bank balance sheet growth and increase in monetary aggregates associated with policy responses to the COVID-19 crisis has reignited debate over whether current central bank actions will cause a spike in inflation if not swiftly reversed.
- Two diametrically opposed views:
  - Some argue central bank balance sheet expansion should be curtailed to short-circuit money growth and inflation.
  - Others, including some associated with “modern monetary theory”, advocate increased reliance on monetary (vs debt) finance during the COVID crisis and for future large-scale public investment, arguing there remains ample scope to employ monetary finance without inflationary consequences.
- The paper reopens the long-running debate over the relation between money and inflation, contrasting the monetarist claim that “inflation is always and everywhere a monetary phenomenon” with the view that monetary quantities play no independent role in determining macro variables.

### Core theoretical contention
- Central point of contention: whether the quantity of money is exogenous (determined by the state: central bank and treasury) or endogenous (determined by private sector demand and supplied passively by the state).
- The state can directly control only the forms of money it issues: currency and deposits held by the private sector at the central bank (variously called “central bank or outside money”, M0, “high powered money”, and the “monetary base”).
- Key empirical operational question: does the state determine the amount of private sector deposits held at the central bank—usually synonymous with bank reserves?

### Historical operational change and its implications
- Prior to the Global Financial Crisis (GFC), central banks generally provided the quantity of reserves demanded by banks, used interest rates rather than quantitative policy instruments, and did not target monetary aggregates.
- Japan is identified as an exception, having started to experiment with balance sheet expansion in 2001.
- Since 2009, in floating exchange rate regimes central bank asset purchases have determined the quantity of the monetary base (making the monetary base “exogenous”), a change that could alter the relationship between money and prices toward a more monetarist view.
- Once central banks reached the effective lower bound on nominal interest rates, they purchased assets in unconventional policy actions and allowed the corresponding excess bank reserves created to remain in the system, so the banking system held substantially more “excess reserves” than had previously been demanded for settlement of interbank payments.

### Motivations for revisiting money–inflation relationship
- Three important motivations articulated:
  - Beliefs matter: If enough people act as if the monetarist view is correct, inflation expectations may rise absent fundamentals, affecting nominal interest rates on newly issued government debt and forcing the state to choose between higher-than-desired inflation or higher-than-desired real debt costs. The paper aims to modestly alter views toward the non-monetarist interpretation of inflation to lessen the probability of adverse equilibria driven by false model expectations.
  - Regime change in money supply determination: The exogeneity of the monetary base since the GFC may have changed the empirical relation between money growth and inflation, warranting re-examination in the new operational reality.
  - Policy relevance regarding monetary finance: If re-examination finds no reason to expect money growth to be correlated with higher inflation, some might conclude that outright money creation is a costless way to finance government spending (e.g., “helicopter money”). The paper argues that developments in modern financial markets invalidate that view and seeks to articulate the key differences between endogenous money growth and exogenous monetary finance and to spell out why and when monetary finance is inflationary.

### Empirical and policy agenda of the paper
- The paper discusses:
  - The modern history of monetarism and its alternatives.
  - The changing empirical relationship of various measures of money and inflation, showing that previous naïve correlations between money and inflation in the 20th century have largely disappeared.
  - Why the disappearance of those correlations cannot be taken as support for increased reliance on permanent monetary finance.
- Anticipated structural forces: rapid technological innovation in payments systems—both public and private—including in global pledged collateral markets, is expected to portend a declining demand for central bank liabilities.

*Source: wpiea2021006-print-pdf — Introduction*

### Section III looks at the changing empirical relationship of various measures of money and

### wpiea2021006-print-pdf - Section III looks at the changing empirical relationship of various measures of money and

### I.  QUANTITY THEORY OF MONEY—THE LAST 50 YEARS (THEORY)
- Key interrelated propositions supporting the Quantity Theory of Money (QTM):
  - (i) the proportionality of money and prices,
  - (ii) the exogeneity of the nominal stock of money and its causal role in the monetary transmission mechanism,
  - (iii) the monetary theory of the price level and
  - (iv) the neutrality of money.
- Historical context and revival:
  - The revival of the QTM is most strongly associated with Milton Friedman.
  - J.M. Keynes: in A Tract on Monetary Reform, “The truth of this, properly explained and qualified, it is foolish to deny”.
  - Important difference between QTM circa 1924 and the Friedman revival: original QTM related money to payments/transactions (not national income); revived QTM related money to expenditures on final goods and services (national income).

### A.  Equation of Exchange
- Definitions and transformations:
  - Original Equation of Exchange (Fisher 1922) expressed as: MV(t) = P(t)T
  - Conventional monetarist formulation transformed into: MV(y)=P(y)Y
  - “Cambridge k” formulation: M = k(y)P(y)Y
  - Velocity (V) defined as turnover rate of money; 1/V interpreted as k in Cambridge approach.
- Conceptual shifts and empirical practicality:
  - Early theory tied money to the value of all economic payments (bond, stock, foreign exchange trading, real estate transactions, tax payments, wholesale and intermediate trade, etc., as well as final goods and services).
  - Pivot to income-based QTM driven largely by data availability and the difficulty of constructing a transactions-based price index.
  - Reinterpretation as a money demand function (Cambridge k) coupled with exogenous money supply to facilitate empirical work.
- Fundamental conceptual problem:
  - Money is used across many markets with different payment systems and velocities; income velocity (ratio of national income to money) is not a meaningful measure of overall money turnover.
  - Example contrasts:
    - Ratio of national income in the United States to currency (primarily used for payments of final goods and services) was about 18.
    - Ratio of U.S. interbank payments to bank reserves (exclusively used in interbank payments) was 77,874 in 2007.
  - Mixing these different market-specific turnover measures is conceptually problematic.
- Distinct views on money endogeneity and dynamics:
  - Monetarists emphasized velocity as technological and relatively stable → changes in money seen as exogenous drivers of income.
  - Keynesians viewed k as a portfolio-balance parameter → nominal income changes drive the quantity of money (money endogenous).
  - Differences affected views on central bank control mechanics (e.g., money multiplier).

### Payments technology, velocity, and empirical consequences
- Technological change and payments:
  - Wholesale payments systems experienced large speed and efficiency gains earlier (fixed costs of innovation spread over high-value payments).
  - Retail payments, apart from currency, also accelerated in advanced countries since the 1980s.
- Consequences for QTM proportionality:
  - Trading in financial markets has grown much faster than the real economy since the 1950s.
  - The discrepancy between money used in retail payments and in total (primarily financial) payments helps explain surprising monetary facts in recent decades.
  - Wholesale payments (through Fedwire, CHIPS, CHAPS, TARGET2) are approximately 50X payments made for final goods and services in the U.S., UK, and Eurozone.
  - Payments for final goods and services comprise less than one percent of total payments.
  - Interpreting a truly transactions-based price index would be extraordinarily difficult.
- Notable empirical statistics and trends (preserved exactly as in source):
  - The ratio of national income in the United States to currency: about 18.
  - The ratio of U.S. interbank payments to bank reserves: 77,874 in 2007.
  - Payments in the main interbank wholesale markets in the U.S., UK, and Eurozone (Fedwire, CHIPS, CHAPS, and TARGET2) are approximately 50X payments made for final goods and services.
  - The sum of payments effected in 2017 through the two U.S. large value transfer systems alone (Fedwire and CHIPS), was 1 quadrillion 133 trillion dollars compared with U.S. GDP that same year of US$19.5 trillion.
  - Conventional income velocity of M1 (GDP divided by M1) was between 3 and 5 from 1927 to 1977.
  - U.S. DD (demand deposit) turnover ratio history:
    - 41 prior to the Great Depression,
    - rises sharply thereafter,
    - 922 in 1996 (the year the Federal Reserve ceased publishing the data).
  - A turnover ratio of 922 on an annual basis implies, given approximately 250 business days per year, that the average deposit balance turns over almost 4 times per day.
  - Total payments made with DD were 58x U.S. GDP in 1996.
  - If one assumes that 85 percent of U.S. GDP was then purchased with debits of U.S. DD (source presents this assumption contextually).

### Implications highlighted in the section
- Stability assumptions underlying QTM undermined by differential technological change across payment systems.
- Income-based QTM may be a practical re-conceptualization but obscures important divergences between transactions in wholesale/financial markets and retail markets.
- Empirical measures of money demand and velocity that aggregate across heterogeneous payment systems can be misleading.
- The rise in wholesale payment efficiency and growth of financial trading implies that traditional monetary aggregates and velocity concepts may not capture the relevant monetary dynamics affecting prices and distributional outcomes.
- Recent debates over macro consequences of ultra-loose monetary policy reflect a partial return to assessing impacts on prices orthogonal to nominal national income (e.g., financial market prices and income distribution), raising questions about central bank targeting beyond retail prices and incomes.

*Source: wpiea2021006-print-pdf - Section III looks at the changing empirical relationship of various measures of money and*

### 1.5 percent of all payments made with DD were then made for the acquisition of final goods

### wpiea2021006-print-pdf - 1.5 percent of all payments made with DD were then made for the acquisition of final goods and services.

### Payments velocities and monetary composition
- 1.5 percent of all payments made with DD were then made for the acquisition of final goods and services.
- Income velocity of currency (C) in the United States has not been rising during the last half of the 20th century, leading to a rising proportion of C in M1.
- C exceeded DD in M1 in the U.S. in 1997 and remained so for twenty years—C was more than double DD just before the GFC in 2007-08.
- The apparent stability of M1 velocity masks diverging velocities: stable or declining velocity of C combined with a relative growth of C in M1 has subdued the income velocity of M1 despite a phenomenal increase in DD transactional velocity.

### Wholesale payments, reserves, and extreme turnover differences
- 1947: US$ 316 billion in payments through Fedwire; stock of reserves at the FRBs US$ 17.9 billion — yielding an annual turnover of 58.
- 2007: US$ 670.7 trillion in payments through Fedwire; reserves amounted to US$ 9.0 billion — yielding an annual turnover of 77,784.
- Put differently, payments increased by 212,115 percent while reserves fell by 50 percent.
- By 2007, C represented about 98 percent of M0 although estimated at most involved in US$ 2 trillion in payments that year while payments made through the two U.S. LVPS, CHIPS and Fedwire, during the same year amounted to US$ 1 quadrillion 156 trillion.
- 2007 average stock of U.S. currency outstanding: US$ 811.7 billion; assuming 50 percent is held outside the United States gives an annual turnover ratio of currency of about 5.
- Average 2007 stock of reserves: US$ 8.6 billion giving a gross LVPS turnover ratio of 134,453 per year.

### Heterogeneous velocities within monetary aggregates and implications for monetarism
- Various subcomponents of M0 and M1 have wildly different “velocities” and stability properties: between C and reserves (M0), and C and DD (M1).
- M0 remains the key monetarist variable as notionally controlled by the central bank and thus exogenous; alternative labels: “base money”, “high-powered” money, “outside” money.
- Currency is widely not rationed by central banks; almost no central bank prevents recipients from returning currency to the banking system.
- Reserves differ from currency: only banks may hold reserves; reserves are assets, not monetary liabilities of banks, and are included only in M0.
- The textbook money multiplier (1/rr) depends on reserve requirements and the ratio C/DD; in reality these relationships have varied considerably over time.

### Endogeneity of reserves and central bank operational practice
- Prior to the GFC advanced-country central banks provided as much currency and reserves as desired at the opportunity cost of a target overnight interest rate; they set the price, not the quantity, of reserves.
- Daylight overdrafts and RTGS operation imply reserves increase during the operating day and fall at the end of the day — the reserves multiplier is unstable and highly variable intraday.
- Example: reported total U.S. funds transfer daylight overdrafts of US$ 76 billion per day compared with total reserve balances around US$ 35 billion (Mengle, Humphrey, Summers (1987) cited).
- Bank-level evidence: 2006 JP Morgan Chase held US$ 1,352 billion in assets of which only US$ 2.2 billion (0.16 percent) were reserves at FRBs.
- Fed’s decision in March 2020 to set the required reserve ratio at zero effectively eliminated reserve requirements for all depository institutions.

### Empirical breakdown of the money–price proportionality (quantity theory tests)
- Re-estimation of McCandless and Weber (MW) methodology across different periods and components reveals departures from proportionality between money growth and inflation, especially for reserves.
- Table 2 (1960–1989): Full Sample (88 countries) coefficients: M0 0.9461, M1 0.9398, M2 0.9299, Currency (C) 0.9604, Reserves (M0-C) 0.6436*, D.Deposits (M1-C) 0.8905*, Other Deposits (M2-M1) 0.9258. (* indicates statistically different from 1 at 5 percent)
- Table 3 (1990–2019): Full Sample (143 countries) coefficients: M0 0.7504*, M1 0.7728*, M2 0.7787*, Currency (C) 0.7841*, Reserves (M0-C) 0.5484*, D.Deposits (M1-C) 0.7250*, Other Deposits (M2-M1) 0.7342*. OECD (20) shows negative coefficient on reserves (-0.3130*).
- Table 4 (2001–2019): Full Sample (140 countries) coefficients: M0 0.6685*, M1 0.72768*, M2 0.8225*, Currency (C) 0.6328*, Reserves (M0-C) 0.4006*, D.Deposits (M1-C) 0.6573*, Other Deposits (M2-M1) 0.7606*. OECD (18) shows reserves -0.3214*.
- Key empirical finding: the coefficient on Reserves (the variable most plausibly exogenous under monetarist logic) is the lowest of all components and often statistically different from 1, including negative in some OECD samples — contradicting a simple causal link from reserves to inflation.

### Japan’s QE experiment and evidence against a reserves→credit→inflation channel
- Bank of Japan QE (2001–2006) targeted current account deposits (reserves) as the operational target.
- Correlations and figures for Japan (1990–2019) show essentially zero correlation between M0 and inflation; change in bank reserves lagged by 12 months and bank credit expansion during 1991-2020 is exactly zero.
- Figures cited: Japan bank reserves and total loans series (1991–2020) show no commensurate impact of reserves expansion on credit.
- Conclusion: expansion in reserves had neither a commensurate impact on credit nor directly on inflation in Japan.

### Post-GFC reserve expansion, capital constraints, and lending
- Superabundant reserves in the United States, UK, Eurozone, Switzerland and Japan have not been associated with marked expansions in credit nor significant inflation increases.
- After the GFC, capital requirements increased and leverage ratios were imposed/raised; capital availability became a more important constraint to lending than reserves.
- In some cases increases in reserves combined with prudential ratios may have paradoxically constrained bank lending.
- Some jurisdictions exempted central bank deposits from leverage ratio calculations (UK), others relaxed ratios in response to COVID-19 (United States).

### Theoretical interpretation: open-market swaps vs “helicopter” money
- Open-market purchases swap treasury securities for reserves; Wallace (1981) shows such swaps are theoretically irrelevant for the economy — reserves versus short-term treasury claims are close substitutes.
- Monetarist “helicopter money” involves money entering the economy without an offsetting asset swap (permanent money finance tied to fiscal transfers), which would plausibly have a stronger inflationary effect.
- The distinction between CBDL created in exchange for bonds and CBDL created to permanently finance fiscal spending is central: the former tends to change asset composition without causing the inflationary outcomes associated with the latter.

### Implications for monetary finance and policy
- Recent large increases in CBDL (reserves and currency) have not caused inflation because much of the increase substituted for financial assets rather than representing permanent, unrequited transfers to the public.
- Advocates of permanent monetary finance who infer non-inflationary outcomes from recent CBDL increases conflate two ways money can enter the economy.
- Unsustainable fiscal expenditures will be inflationary regardless of whether financed initially with money or bonds.
- Excess reserves beyond payment-system needs are “deadweight” on banks’ balance sheets: reserves can only be held by banks, are poor collateral, and are subject to capital requirements; treasury securities are more efficient and tradeable widely.
- Historical development of domestic debt markets in many countries makes debt issuance a preferred way to finance deficits relative to returning to broad monetary finance.

### Conclusions and policy takeaways
- Payments innovation increases speed and reduces the quantity of settlement balances needed; faster, safer payments imply less CBDL required for a given level of transactions.
- Three fundamental points:
  - (i) The relationship between the quantity of CBDL and inflation has completely broken down and shows no causal relationship in modern data.
  - (ii) Recent exogenous increases in CBDL (policy-driven) have not altered (i) and have made the lack of causal linkage clearer.
  - (iii) Permanent monetary finance should not be the preferred route in the near-zero (or negative) rate environment.
- Monetary aggregates are tangential to central banks’ objectives in practice; interest-rate policy remains the operational focus for most central banks.
- Given secular declines in demand for CBDL and the technological evolution of payments, policy relying on increased permanent monetary finance runs counter to historical and technological trends.

*Source: wpiea2021006-print-pdf - 1.5 percent of all payments made with DD were then made for the acquisition of final goods and services (IMF PDF content provided).*

### Annex 1. Collateral and Money Aggregates

### Annex 1. Collateral and Money Aggregates

### Role of collateral in money markets and monetary policy transmission
- Collateral plays a central but often overlooked role in money markets and market plumbing (repo, securities lending, prime brokerage financing, derivatives margins).
- Quantitative easing (QE) removes good collateral—typically sovereign bonds, but even corporates and equities—from the market, with implications for associated rates and market functioning.
- QE and large central bank balance sheet actions mean central banks continue to play a major role in collateral markets.

### Augmenting the IS-LM framework to include collateral
- Traditional LM specification: M = f(Y, r), where M is money supply, Y is output, and r is benchmark interest rates.
- Proposed enriched specification: M = f(Y, r, c_p), where c_p measures the efficiency of pledged collateral (that can be reused).
- Interpretation of c_p:
  - c_p is a parameter governing the efficiency of monetary policy transmission given collateral’s lubricant role in secured finance markets.
  - When collateral markets are constrained (c_p declines), for a given M a larger change in r is required to produce the same change in Y.
  - All else equal, a shrinkage in collateral supply reduces the efficiency of monetary policy transmission.

### Dynamics of LM under collateral constraints and QE
- In the ‘new’ IS-LM model:
  - Changes in monetary policy may not produce a parallel shift in the LM curve; the LM curve can pivot and intersect IS at different points depending on slope and c_p.
  - This framework explains why QE may increase output initially but have diminishing effects as QE increases in scale (consistent with referenced recent research).
- Empirical/market examples:
  - After the Lehman crisis, pledged collateral volumes (including price effects) fell from around US$10 trillion before the crisis to around US$5.5 trillion afterwards; volumes are inching back slowly to US$8.5 trillion. Overall financial collateral efficiency, c_p, declined considerably as a result.
  - QE can cause a rightward shift in the LM curve (via higher M) while simultaneously lowering c_p by removing vast amounts of treasury securities and other good collateral—producing countervailing effects on output and rates.
  - Market dysfunction episodes: dealers shied away from making a market in U.S. Treasury repo in September 2019 and avoided making a market in off-the-run U.S. Treasuries in March 2020, underscoring dealer balance sheet constraints that limit pledged collateral “reuse” rate even with large debt issuance.

### Transmission effects and market plumbing implications
- When collateral use drops:
  - Financial intermediation slows, analogous to drying up of interbank markets.
  - Investors concerned about counterparty risk may lend fewer securities, leaving idle collateral in segregated accounts.
  - A widening of eligible asset pools (increasing pledgeability) or large-scale QE (draining good collateral) both alter c_p and market functioning.
- Consequences for associated rates and market segments include impacts on repo rates, securities lending, prime brokerage financing, and derivatives margins.

### Policy implications and trade-offs
- The enriched IS-LM model implies important policy considerations:
  - Policymakers should recognize the trade-off between the negative effects of constraining collateral markets (lower c_p, impaired transmission, market plumbing stress) and the positive effects of QE (higher M and supportive output effects).
  - With interest rates near zero and QE deployed in response to Covid-19, attention to collateral market functioning is crucial when charting policy paths beyond traditional money aggregates.
- Operational constraints matter:
  - Rebound in pledged collateral reuse rates depends on dealer balance sheet space; large debt issuance alone may not restore market plumbing without addressing balance-sheet capacity.

*Source: Annex 1. Collateral and Money Aggregates (wpiea2021006-print-pdf).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021006-print-pdf.pdf_