## Central Bank Digital Currencies and Financial Stability: Balance Sheet Analysis and Policy Choices

## Source details

**Canonical URL:** [Central Bank Digital Currencies and Financial Stability: Balance Sheet Analysis and Policy Choices](https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024226-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2024/english/wpiea2024226-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2024/english/wpiea2024226-print-pdf.pdf.json)

---

### Executive summary: purpose, scope, and key drivers
- Purpose:
  - Evaluates the financial stability implications of the issuance of retail central bank digital currencies (CBDCs) to inform policy choices to mitigate potential adverse impacts.
- Scope and baseline assumptions:
  - CBDC issued against other central bank liabilities (either cash or reserves).
  - Positive demand for CBDC.
  - Issuance of CBDC occurs through the banking system.
- Key drivers of financial-stability implications:
  - The size of the issuance.
  - Initial conditions of sectoral balance sheets (cash, deposits, excess reserves).
  - Reactions of the banking sector and the central bank.
- Timing and amplification:
  - Adverse effects can arise in steady state but are larger during rapid adoption (network effects) and in crises (risk of digital runs into CBDC).

### Scenario analysis: structure, mechanics, and qualitative conclusions
- Scenario taxonomy (preserved features):
  - Scenario 0: CBDC replaces cash; Volume of deposit outflow: Zero; Reserve offset mechanism: No; Second round effects: No.
  - Scenario 1: CBDC replaces deposits; Volume of deposit outflow: Small; Excess reserves: Sufficient; Second round effects: No.
  - Scenario 2-A: Deposits; Volume of deposit outflow: Medium; Excess reserves: Insufficient; Reserve offset mechanism: CB lending; Second round effects: No.
  - Scenario 2-B: Deposits; Volume of deposit outflow: Large; Excess reserves: Insufficient; Reserve offset mechanism: CB lending (balance sheet expansion); Second round effects: No.
  - Scenario 3-A: Deposits; Volume of deposit outflow: Large; Excess reserves: Insufficient; Reserve offset mechanism: Domestic borrowing; Second round effects: No.
  - Scenario 3-B: Deposits; Volume of deposit outflow: Large; Excess reserves: Insufficient; Reserve offset mechanism: Foreign borrowing; Second round effects: No.
  - Scenario 4-A: Deposits; Volume of deposit outflow: Large; Reserve offset mechanism: Loan shrinks, CB lending; Second round effects: Yes (lending rates).
  - Scenario 4-B: Deposits; Volume of deposit outflow: Large; Reserve offset mechanism: Loan shrinks, CB lending; Second round effects: Yes (deposits and lending rates).
- Core mechanics (Box 1 summary):
  - CBDC purchased by households reducing deposits → reduction in banks’ reserves unless central bank generates additional reserves via lending or asset purchases.
  - Total reserves change only through central bank transactions (open market operations, lending) or autonomous factors.
  - Ways to generate reserves:
    - Central bank lending (temporary).
    - Asset purchases (permanent).
  - Different reserve-generation methods imply different banking-sector adjustments (increased central bank borrowing, increased wholesale funding from NBFIs or foreign sources, or contraction of bank assets/loans).

### Quantitative illustrative outcomes: prudential and profitability metrics (values preserved)
- Basel III and profitability metrics across baseline and scenarios (values shown exactly as in source):
  - LR: Baseline 0.050; Scenario 0 0.050; Scenario 1 0.050; Scenario 2-A 0.050; Scenario 2-B 0.048; Scenario 3-A 0.050; Scenario 3-B 0.050; Scenario 4-A 0.051; Scenario 4-B 0.053
  - LCR: Baseline 1.40; Scenario 0 1.40; Scenario 1 1.39; Scenario 2-A 0.65; Scenario 2-B 0.57; Scenario 3-A 1.12; Scenario 3-B 0.90; Scenario 4-A 0.44; Scenario 4-B 0.70
  - NSFR: Baseline 1.09; Scenario 0 1.09; Scenario 1 1.09; Scenario 2-A 0.98; Scenario 2-B 0.92; Scenario 3-A 0.95; Scenario 3-B 0.92; Scenario 4-A 0.95; Scenario 4-B 0.99
  - NIM: Baseline 0.0278; Scenario 0 0.0278; Scenario 1 0.0275; Scenario 2-A 0.0264; Scenario 2-B 0.0248; Scenario 3-A 0.0243; Scenario 3-B 0.0256; Scenario 4-A 0.0273; Scenario 4-B 0.0256
  - ROE: Baseline 0.087; Scenario 0 0.087; Scenario 1 0.084; Scenario 2-A 0.053; Scenario 2-B 0.043; Scenario 3-A 0.009; Scenario 3-B 0.033; Scenario 4-A 0.055; Scenario 4-B 0.010
  - ΔLoans: Baseline (blank); Scenario 0 0; Scenario 1 0; Scenario 2-A 0; Scenario 2-B 0; Scenario 3-A 0; Scenario 3-B 0; Scenario 4-A -4%; Scenario 4-B -8%
- Key quantitative illustrations (selected baseline and scenario calibrations, values preserved):
  - Baseline consolidated banking sector balance sheet size: 10,000 (arbitrary unit).
    - Loans: 6,000 (60 percent of total assets).
    - Government bonds: 1,500 (15 percent).
    - Reserves: 140 (1.4 percent).
    - Others: 2,360 (23.6 percent).
    - Deposits (liabilities): 7,000 (70 percent of total liabilities).
    - Lending from the central bank: 750 (7.5 percent).
    - Debt issued: 500 (5 percent).
    - Capital: 500 (5 percent).
  - Central bank balance sheet: 2,000 (one fifth of banking sector).
    - Central bank assets: securities 1,000; lending to commercial banks 750; others 250.
    - Central bank liabilities: cash 1,000; commercial bank reserves 140; others 860.
  - Scenario calibrations:
    - Scenario 0 (cash-like CBDC): CBDC issuance assumed at 400 (5 percent of value of money used by households).
    - Scenario 1 (excess reserves absorb CBDC): initial reserves assumed to be 600; required reserves at 2 percent of deposits equal 148; CBDC issuance: 400.
    - Scenario 2-A: CBDC replaces deposits by 750 (around 9 percent share in households’ liquid assets).
    - Scenario 2-B: CBDC replaces 1,200 of deposits (15 percent of households’ liquid assets).
    - Scenario 3-B and Scenario 4: CBDC issuance calibrated at 1,200 in illustrative tables.

### Main financial-stability implications and channels
- When CBDC largely substitutes physical cash:
  - No material financial stability implications; central bank and banking system size largely unchanged (Scenario 0).
- When CBDC competes with deposits:
  - If banking sector has ample excess reserves (Scenario 1), impacts are contained (reserves drawn down).
  - If excess reserves are insufficient, central bank asset expansion or banks’ reliance on wholesale/foreign funding can arise (Scenarios 2, 3, 4), producing material effects:
    - LCR and NSFR can fall below regulatory minima (Scenarios 2-A, 2-B, 3-B).
    - Leverage Ratio can decline where banks expand balance sheets to obtain collateral (Scenario 2-B).
    - Profitability (NIM, ROE) tends to fall as cheap deposit funding is replaced by more costly funding sources.
    - Second-round effects (Scenario 4-A, 4-B) transmit higher funding costs to loan rates and credit contraction: Scenario 4-A ΔLoans -4%; Scenario 4-B ΔLoans -8%.
- Additional systemic channels:
  - Increased reliance on wholesale funding strengthens interconnections with NBFIs and foreign counterparties (currency and maturity mismatch risks in Scenario 3-B).
  - Sovereign–bank nexus can strengthen as collateral demand for sovereign bonds rises.
  - Loss of deposit-based payment data can deteriorate banks’ credit assessment capacity unless CBDC distribution/participation designs preserve information flows (Box 3).
  - Digital runs: widely accessible CBDC can act as “safe haven,” accelerating runs on individual banks or the system in crises.

### Policy options analyzed and their tradeoffs
- Broad categories:
  - CBDC design choices (primary mitigation channel).
  - Macroprudential and financial-sector regulatory policies.
  - Central bank lending operations and monetary policy adjustments.
  - Private payment arrangements and interoperability measures.
- CBDC design (most promise):
  - Encourage CBDC use as means of payment (extensive margin) while inducing small individual holdings (intensive margin).
  - Design tools: holding caps (aggregate or per-wallet limits), zero- or negative remuneration, and high interoperability with bank-operated payment systems (instant convertibility, bank-distributed wallets, on-demand creation/retirement).
  - Specific empirical indications in source: Bank of England proposals between £10,000 and £20,000 per individual; ECB considers around 3,000 to 4,000 digital euros per capita (design discussion preserved as in source).
- Macroprudential policy:
  - Measures that may be warranted:
    - Hike banks’ capital buffers ahead of launch.
    - Caps on FX funding or applying LCR requirements separately by major funding currency.
    - Stressed debt service-to-income ratios.
  - Limitations:
    - Macroprudential tools improve resilience but likely insufficient alone to fully offset liquidity pressure, deposit runs, or credit intermediation disruption.
    - Progress needed on macroprudential regulation of NBFIs, but complexity and cross-border nature make this challenging.
- Central bank market operations and LOLR:
  - Expanding central bank lending can supply reserves but has limits:
    1. Central bank funding typically more expensive than deposit funding → raises banks’ funding costs, reducing profitability and lending.
    2. Collateral constraints: broadening collateral transfers credit risk to central bank and may not improve unencumbered HQLA for LCR.
    3. Typical short-term nature of central bank lending does not improve NSFR.
  - Longer-tenor lending (LTRO/TLTRO) can help in crises but is unattractive in steady state as it displaces maturity transformation by banks.
  - Enhancements for crisis response:
    - Ex-ante positioning of illiquid assets and pre-specified haircuts to widen eligible collateral in emergencies.
    - Forward-looking solvency assessments and possible government guarantees to facilitate temporary support to banks.
  - Trade-offs: expanding LOLR capacity implies operational and balance-sheet risks to central bank.
- Other regulatory and supervisory measures:
  - Strengthen banking-sector capital, supervision, resolution frameworks, deposit insurance (sufficient coverage and fast compensation) before CBDC launch.
  - Strengthen cyber resilience and fraud prevention; preserve physical cash as a backup.

### Design and market arrangements that preserve payments efficiency while containing risks
- Promote tokenized deposits and private deposit-based payment solutions that settle in central bank money (wholesale CBDC or settlement mechanisms) to preserve the singleness of money and avoid adverse effects on credit intermediation.
- Interoperability options that minimize outstanding CBDC balances (on-demand creation/retirement or instant convertible wallets) reduce CBDC’s attractiveness as a store of value.

### Conclusions and policy guidance (condensed)
- Material adverse financial stability effects depend on:
  - issuance size,
  - initial conditions (cash prevalence, excess reserves, banks’ collateral holdings),
  - reactions of banks and central bank.
- Policymakers can contain detrimental effects by:
  - adopting CBDC designs that encourage use as a means of payment rather than a store of value (caps, zero/negative remuneration, interoperability);
  - using central bank lending and collateral policy carefully, particularly in crisis times (prepositioning assets, specifying haircuts);
  - pursuing macroprudential and regulatory measures while recognizing their limits, particularly given NBFI complexity.
- Key diagnostic indicators to assess likely scenarios include:
  - Cash-in-circulation-to-GDP and transfer-deposits share in banking liabilities (likelihood of CBDC replacing cash vs deposits).
  - Share of reserves in banks’ balance sheets (capacity to absorb deposit-to-CBDC shifts).
  - Interest margins (degree of pass-through and risk of credit contraction).

*Source: wpiea2024226-print-pdf — IMF Working Paper (Executive Summary, Introduction, Scenario analysis, Boxes 1–4, Policy options, and Conclusion).*

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

### Executive Summary

### Key purpose and scope
- Evaluates the financial stability implications of the issuance of retail central bank digital currencies (CBDCs) to help inform policy choices to mitigate potential adverse impacts.
- Emphasizes that financial stability implications of issuing a retail CBDC depend on (1) the size of the issuance, (2) initial conditions, and (3) the reactions of the banking sector and the central bank.

### Scenario analysis and main drivers of impact
- Provides a range of scenarios tracing how balance sheets of the commercial banking system and of the central bank may respond to issuance of digital money, affecting:
  - sources and stability of funding;
  - profitability; and
  - provision of credit of the financial system.
- Highlights that adverse effects on the banking system become larger when:
  - (1) the replacement of deposits is greater;
  - (2) commercial banks do not hold excess central bank reserves; and
  - (3) central banks are not willing to increase the supply of reserves beyond existing collateral requirements.
- Notes that adverse implications can arise in steady state, but may be more material during:
  - rapid adoption driven by network effects, and
  - crisis times when there is a risk of digital runs into the CBDC.

### When financial stability concerns do and do not arise
- No material financial stability implications when the CBDC largely replaces physical cash.
- If CBDC competes with deposits:
  - implications are largely contained if the banking sector holds ample excess reserves that can be drawn down in exchange for the CBDC;
  - adverse effects mostly arise where the central bank balance sheet expands and commercial banks increase more costly and/or more volatile funding to generate additional reserves.

### Policy options analyzed
- Examines a comprehensive set of policy options to mitigate adverse effects on financial stability, including:
  - macroprudential policy tools;
  - expansion in central bank lending to the commercial banking system, including changes to central bank collateral policies; and
  - CBDC design choices to mitigate adverse financial stability implications.

### Findings on policy effectiveness and tradeoffs
- Macroprudential and financial sector regulatory policies:
  - Can increase banks’ resilience but are likely insufficient alone if CBDC puts pressure on bank liquidity or disrupts credit intermediation.
  - Financial stability concerns from a greater role of non-bank financial institutions (NBFIs) in banks’ funding underscore the need to make progress in macroprudential regulation of NBFIs, but the diversity and international nature of some NBFIs make this complex.
- Central bank lending operations:
  - Expanding central bank lending can help reduce detrimental financial stability impacts, particularly in crisis times when emergency liquidity to banks facing runs into the CBDC may be required.
  - Greater use of ex-ante positioning of illiquid assets and applying haircuts on these assets in emergency operations is worth considering.
  - Loosening central bank lending standards in normal times or steady state (for example, by admitting a broader set of collateral or aiming for longer loan tenors in open market operations) can expose central bank balance sheets to substantial additional risks.
- CBDC design features hold the most promise for alleviating financial stability concerns:
  - Designs that encourage use of the CBDC as a means of payment, rather than as a store of value, are likely to maintain benefits while avoiding costs for financial stability.
  - Design elements cited include zero- or negative remuneration, limits to holding CBDC, and a high degree of interoperability with the bank-operated payment system.
  - These features can ensure CBDC is used primarily for transactions, while containing risks to credit disintermediation as CBDC adoption advances.
  - CBDC designs that foster further developments in deposit-based payment solutions — such as tokenized deposits — could also preserve the singleness of money while containing negative effects on financial stability.

### Conclusion
- Material adverse financial stability effects depend on issuance size, initial conditions, and reactions of banks and the central bank.
- Policy makers can contain detrimental effects by:
  - adopting CBDC designs that encourage use as a means of payment rather than a store of value;
  - using central bank lending and collateral policy carefully, especially in crisis times; and
  - pursuing macroprudential and financial sector regulatory measures while recognizing their limits, particularly given NBFI complexity.

*IMF Working Paper — Executive Summary*

### 1.     Introduction

### 1.     Introduction

### Context and motivation
- Central banks worldwide are exploring potential benefits of CBDCs with objectives ranging from modernizing the payments system and making it more resilient, to ensuring trust in the monetary system and monetary sovereignty (Soderberg et al., 2022).
- According to Kosse and Mattei (2023), over the course of 2022, more than 90 percent of central banks engaged in some form of CBDC work, and around two thirds of central banks considered that they were likely to or might possibly issue a retail CBDC in either the short or medium term.
- Central banks and ministries of finance have established the principle that potential issuance of CBDC should “do no harm” to the existing financial system (Bank for International Settlements (BIS) Innovation Hub (2021a), G7 (2021)). “No harm” is not meant to be interpreted as “no impact” (BIS 2021a).

### Contribution and analytical approach
- Gap addressed: existing literature lacks a full analysis of potential impacts on the financial sector as a whole, including the banking sector, the central bank, and nonbank financial institutions (NBFIs).
- Framework: sectoral balance sheet analysis to assess implications of a retail CBDC for financial stability.
- Three baseline assumptions used to construct scenarios:
  - (1) central banks issue CBDC against other central bank liabilities (either cash or reserves) rather than by way of a “helicopter drop,”
  - (2) there is positive demand for CBDC, and
  - (3) issuance of CBDC occurs through the banking system.
- The analysis provides a systematic account of potential balance sheet adjustments for both the central bank and commercial banks and can be applied across a range of country starting points and CBDC objectives.

### Key drivers of adverse implications
- Potential adverse implications for the existing financial system depend on:
  - (1) the amount of CBDC being issued,
  - (2) the preconditions relating to the existing sectoral balance sheets prior to the issuance of the CBDC, and
  - (3) basic assumptions on how banks and the central bank interact in the issuance of CBDC.
- Adverse effects can be magnified by:
  - rapid adoption exceeding expectations, and
  - crisis scenarios when banks struggle to replace funding lost in a digital run.

### Main findings on banking system impacts
- When issuance is against existing physical cash, or leads to a drawdown of existing excess reserves, implications for the existing financial system are generally likely to be mild.
- When issuance leads to an expansion of the central bank’s balance sheet, adverse effects can arise if banks resort to more costly or volatile wholesale funding from the central bank or NBFIs.
- Banks’ competitive reactions to CBDC issuance can increase rates charged on loans to households and firms, ultimately reducing the volume of credit to households and firms.
- Some adverse effects can occur in a new steady state where CBDC coexists with traditional deposits; these effects can be larger during rapid adoption or crises.
- Greater interconnectedness with the NBFI sector and the foreign sector introduces new financial stability risks.

### Policy mitigation overview
- Main categories of policy response:
  - macroprudential policy,
  - expansion in provision of central bank liquidity to the banking system,
  - design options for the CBDC to reduce its role as a store of value.
- Design-related mitigants:
  - Encourage CBDC use as a means of payment (extensive margin) while inducing households and firms to hold small amounts for retail payments (intensive margin).
  - Design elements include quantitative restrictions (caps), maintaining zero remuneration of CBDC balances, and ensuring interoperability with bank deposits.
  - Note: quantitative restrictions can work better when CBDC is not remunerated to limit circumvention via multiple wallets.
- Macroprudential measures:
  - Can increase banks’ resilience (for example, stronger capital positions ahead of CBDC introduction).
  - Mitigating emerging risks by macroprudential or regulatory policies alone is likely to be difficult because CBDC introduction may erode prudential liquidity metrics that tightening requirements cannot easily address.
  - Urgency of progress on macroprudential regulation of NBFIs, though this is likely to be challenging.
- Central bank liquidity provision:
  - Expansion of lending operations can help reduce detrimental financial stability impacts, especially in crisis times to provide emergency liquidity to banks subject to runs into the CBDC.
  - Changes to liquidity provision in normal times (broader collateral, longer tenors) can expose central bank balance sheets to additional risks.

### Scope and exclusions
- Focus: implications of issuing a retail CBDC (available for everyday payments by households and firms).
- Excluded from scope:
  - formal welfare analysis,
  - effects on the effectiveness of monetary policy,
  - comprehensive treatment of financial stability concerns related to private digital money that may motivate CBDC issuance,
  - detailed discussion of fast payment initiatives and other alternatives (these are discussed elsewhere, e.g., Das et al., 2023).
- The paper assumes the central bank maintains the plan to issue a retail CBDC and seeks to reduce associated adverse effects.

### Roadmap of the chapter
- Section 2: summarizes the range of benefits of issuing a CBDC for central bank objectives.
- Section 3: presents analysis of CBDC adoption scenarios using balance sheet traces.
- Section 4: discusses adverse financial implications from scenarios.
- Section 5: examines risk-mitigating policies, including macroprudential policies, monetary policy and central bank lending operations, and CBDC design options to limit store-of-value use.
- Section 6: concludes with policy implications.

*Source: wpiea2024226-print-pdf - 1.     Introduction*

### 3.     CBDC Issuance Scenarios: Impact on

### 3.     CBDC Issuance Scenarios: Impact on Financial Sector Balance Sheets

### Overview: scope and objectives
- Consider effects of CBDC issuance on central bank and banking sector balance sheets under various scenarios reflecting uncertainty about future CBDC demand.
- Volume demanded by households is an important determinant of financial stability effects.
- Analysis focuses on how effects depend on:
  - preconditions (structure of central bank and commercial bank balance sheets),
  - second-round effects from competitive reactions by the commercial banking sector,
  - endogenous reactions by agents (households, firms, banks) to the new central bank liability.
- Key insight: although some scenarios anticipate significant adverse implications for the banking sector, detrimental effects may be contained in many economies (examples given):
  - economies with sizable amounts of physical cash: retail CBDC may substitute physical for digital cash with very mild, if any, adverse impacts on financial intermediation;
  - economies where the banking sector carries large amounts of excess reserves: CBDC issuance would reduce excess reserves without banks needing to replace lost deposits.

### Baseline Scenario
- Uses illustrative values (in an arbitrary unit) of balance sheet components for the consolidated banking sector and the central bank.
- Consolidated banking sector assets: loans to the private sector, reserves held with the central bank, government bonds, and other securities.
  - Government bonds are distinguished from other securities because they are typically the only securities usable as collateral by banks when borrowing from the central bank.14
- Banks are subject to minimum reserve requirements, determining excess reserves held on deposits with the central bank.
- Consolidated banking sector liabilities: deposits, central bank lending, debt issued, other short-term liabilities, and capital.
  - Deposits are the primary and cheapest funding source, representing most of banks’ total liabilities.
- Central bank balance sheet typically smaller than the commercial banking system:
  - Central bank liabilities: cash and commercial bank reserves as main liabilities.
  - Central bank assets: outright securities holdings and lending to the commercial banking system as main assets.
- Table 1 (referenced) describes the balance sheet in the Baseline Scenario (without CBDC); details in Annex I.

### CBDC Replacement Scenarios: structure and assumptions
- Stylized scenarios trace balance sheet changes for central bank and commercial banking system resulting from CBDC introduction.15
- Differences in issuance volumes reflect variations in consumer demand for CBDC, taken as given in scenario impacts.
- Scenario 0: CBDC replaces cash.
  - Households hand over an equal amount of cash to the central bank (or to a bank acting on behalf of the household) to obtain CBDC.
  - Reserves are unaffected in Scenario 0.
- Other scenarios (Scenarios 1–4 and subvariants):
  - Households draw down commercial bank deposits to purchase CBDC.
  - Households’ bank buys CBDC on behalf of customers from the central bank and pays using its existing reserves at the central bank.
  - These scenarios lead to deposit reductions in the banking system and reductions in reserves (similar to households increasing cash holdings by reducing deposits).
  - Central bank may need to restore or generate additional reserves unless the system holds excess reserves and the central bank accommodates the decrease in reserves (Scenario 1 considers this case).
- Ways for central bank to generate additional reserves:
  - Increase its own assets via additional central bank lending or purchase of bonds (open market operations).
  - These reserves-generating transactions neutralize the reduction in reserves arising from CBDC issuance.
  - Different reserve generation methods have different implications for how banks compensate for deposit declines (e.g., increased borrowing from the central bank, increased borrowing from domestic non-banks or foreign banks).
- When moving across scenarios, CBDC demanded (and deposits withdrawn) increases, requiring larger central bank transactions to generate required reserves.
- Three modeling assumptions in scenarios:
  1. central banks issue CBDC against other central bank liabilities (either cash or reserves);
  2. there is positive demand for CBDC;
  3. issuance of CBDC occurs through the banking system.
- Which scenario materializes depends on:
  - volume of CBDC demanded,
  - amount of cash and deposits in the system,
  - quantity of excess reserves,
  - reserve-offset mechanisms used by the central bank.
- Empirical context examples:
  - In countries where cash is predominant (certain African and Caribbean countries), issuance may primarily drain cash → Scenario 0 likely.
  - In economies where cash is less common and substitution is with bank deposits, issuance primarily reduces deposits → draining of reserves → Scenario 1 if banking system holds excess reserves (e.g., several advanced economies after QE post-GFC).
  - In absence of excess reserves, central bank must counterbalance reduction, leading to progressively stronger effects in Scenarios 2, 3, and 4.
- All scenarios assume net sum of assets in the domestic economy remains unaltered:
  - CBDC does not create new “net wealth” and is not introduced via a “helicopter drop”; it is issued in exchange for an existing asset (cash or commercial bank reserves).

### Box 1: Changes in balance sheets following CBDC issuance through reserves and open market operations
- Total reserves in the banking system are determined exclusively by transactions with the central bank.
  - Interbank reserve transfers do not change total reserves; only open market operations or changes in autonomous factors (issuance of banknotes, government deposits) change total reserves.
- When households increase cash holdings by withdrawing deposits, reserves decline because banks use reserves to purchase banknotes.
- Similar effect when households purchase CBDC by reducing deposits: CBDC is a direct central bank liability and issuance reduces banks’ reserves.
- Table A-1 (referenced) summarizes changes in central bank, banks, and households balance sheets:
  - Central bank: Assets and Liabilities changes listed include Reserves and CBDC.
  - Banks: Reserves (asset) and Deposits (liability) change.
  - Households: Deposits and CBDC change.
- Reduction in banks’ reserves implies, all else equal, higher banks’ demand for reserves and upward pressure on interbank rates.
  - Banks may demand more reserves to satisfy reserve requirements if holdings were close to requirements prior to CBDC introduction.
- Central bank responses:
  - Adjust reserve requirements, operate a floor system (supply abundant reserves at desired interest rate), or use open market operations to increase reserve supply.
  - Open market operations can be temporary (central bank lending guaranteed by collateral; banks repay after a period) or permanent (asset purchases such as government bonds implying a permanent increase in total reserves).
- Table A-2: balance sheet changes when reserves are generated through central bank lending → corresponds to Scenario 2-A.
  - Banking sector’s balance sheet size does not change.
  - Central bank’s balance sheet expands due to CBDC issuance.
- Table A-3: (referenced) summarizes balance sheet changes following permanent operations (asset purchases).

### Summary of Each Scenario (qualitative findings)
- Scenario 0:
  - CBDC distributed in exchange for cash.
  - CBDC assumed to be a close substitute of cash.
  - CBDC mainly drains cash from the economy; impact on deposits is very limited.
  - Relevant for economies where cash use predominates.
- Scenario 1:
  - CBDC replaces deposits in households’ balance sheets.
  - Banks are assumed to possess excess reserves.
  - When households demand CBDC, the commercial banking system draws down existing excess reserves in exchange for central bank supplying the CBDC.

*Source: IMF Working Paper chapter titled "3.     CBDC Issuance Scenarios: Impact on Financial Sector Balance Sheets" from the provided PDF content.*

### Box 1. Changes in Balance Sheets Following CBDC Issuance Through

### Box 1. Changes in Balance Sheets Following CBDC Issuance Through

### Mechanisms of balance sheet adjustment and asset purchase operations
- In central bank asset purchase operations, banks sell government bonds to the central bank. If banks do not hold enough government bonds, NBFIs instead sell government bonds to the central bank, and then lend the additional reserves to the commercial banking sector.
- Sequence when NBFIs sell government bonds to the central bank:
  - Step 1: NBFIs holding reserve accounts sell government bonds (GB) to the central bank and increase reserve holdings (described in Table A-3).
  - Step 2: NBFIs lend these additional reserves to the commercial banking sector via wholesale funding (wholesale lending), enabling banks to obtain liquidity.
- If NBFIs do not hold reserve accounts directly, transactions can be settled through custodian banks that hold reserve accounts (example: pension funds selling government bonds to the central bank settle via custodian banks).

### Scenario taxonomy and distinguishing features
- Table 2 summarizes scenario features along four dimensions: (1) whether CBDC replaces cash or deposits, (2) volume of deposit outflow, (3) whether excess reserves are sufficient, (4) reserve offset mechanism, and (5) whether second-round effects are considered.
- Scenarios (as described):
  - Scenario 0
    - CBDC Replacement: Cash
    - Volume of deposit outflow: Zero
    - Excess reserves: (blank)
    - Reserve offset mechanism: No
    - Second round effects: No
  - Scenario 1
    - CBDC Replacement: Deposits
    - Volume of deposit outflow: Small
    - Excess reserves: Sufficient
    - Reserve offset mechanism: (blank)
    - Second round effects: No
  - Scenario 2-A
    - CBDC Replacement: Deposits
    - Volume of deposit outflow: Medium
    - Excess reserves: Insufficient
    - Reserve offset mechanism: CB lending
    - Second round effects: No
  - Scenario 2-B
    - CBDC Replacement: Deposits
    - Volume of deposit outflow: Large
    - Excess reserves: Insufficient
    - Reserve offset mechanism: CB lending (balance sheet expansion)
    - Second round effects: No
  - Scenario 3-A
    - CBDC Replacement: Deposits
    - Volume of deposit outflow: Large
    - Excess reserves: Insufficient
    - Reserve offset mechanism: Domestic borrowing
    - Second round effects: No
  - Scenario 3-B
    - CBDC Replacement: Deposits
    - Volume of deposit outflow: Large
    - Excess reserves: Insufficient
    - Reserve offset mechanism: Foreign borrowing
    - Second round effects: No
  - Scenario 4-A
    - CBDC Replacement: Deposits
    - Volume of deposit outflow: Large
    - Excess reserves: Insufficient
    - Reserve offset mechanism: Loan shrinks, CB lending
    - Second round effects: Yes (lending rates)
  - Scenario 4-B
    - CBDC Replacement: Deposits
    - Volume of deposit outflow: Large
    - Excess reserves: Insufficient
    - Reserve offset mechanism: Loan shrinks, CB lending
    - Second round effects: Yes (deposits and lending rates)

### Detailed scenario mechanics and second-round effects
- Scenario 2-A
  - Banks hold only required reserves; requirement unchanged by central bank issuance of CBDC.
  - Banks hold sufficient government bonds to pledge with the central bank to obtain additional reserves needed for purchase of CBDC via CB lending (standard open market operation).
  - Amount of loans to households and firms can initially stay the same.
- Scenario 2-B
  - CBDC replaces deposits to a greater extent (e.g., attractive CBDC design such as a positive interest rate).
  - Banks lack sufficient government bonds to access central bank lending fully.
  - Banks first increase wholesale funding from NBFIs to purchase government bonds from NBFIs, then use these bonds as collateral to borrow reserves from the central bank.
  - Commercial banks’ balance sheets expand, while loans initially remain unchanged.
- Scenario 3-A
  - Central bank purchases additional government securities from the NBFI sector, generating an inflow of reserves into the banking system in exchange for an increase in wholesale funding sourced from NBFIs.
  - This wholesale funding replaces lost deposit funding, keeping commercial banks’ balance sheet size unchanged.
  - Wholesale funding is assumed to be offered by NBFIs because all banks face a fall in deposits and reserves.
- Scenario 3-B
  - Domestic NBFIs unwilling or unable to provide wholesale funding or to sell government bond holdings to the central bank.
  - Banks rely on foreign borrowing in the form of short-term funding, while the central bank purchases foreign government bonds rather than domestic ones.
  - Overall size of the domestic banking system remains unchanged.
- Scenario 4-A (second-round effect)
  - Higher funding costs on commercial banks’ balance sheets (from additional CB funding or additional wholesale funding).
  - Banks partially pass on higher funding costs to loan rates, reducing loans to households and firms compared to baseline and previous scenarios.
- Scenario 4-B (additional second-round effect)
  - Commercial banking system increases remuneration of deposits to stem deposit outflow.
  - Competitive response increases funding costs across the deposit base and costs to plug outflow into CBDC.
  - Stronger adverse impacts on loans than in Scenario 4-A; loans decline more due to higher overall funding costs.

### Changes in central bank balance sheet across scenarios
- Two broad outcomes for central bank balance sheet:
  - No change in overall size (Scenarios 0 and 1): central bank decreases other liabilities when issuing CBDC (switch between CBDC and cash or allow reduction in reserves due to ample excess reserves).
  - Expansion of assets (Scenarios 2, 3, and 4): central bank increases assets to match increased liabilities from CBDC issuance. Options include:
    - Boost lending to commercial banks (Scenarios 2-A and 2-B).
    - Increase holdings of securities (Scenarios 3-A and 3-B).
    - In Scenario 4 the central bank is assumed to increase both lending and securities holdings.
- Note: The central bank controls which type of asset holdings it expands in Scenarios 2, 3, and 4.

### Tracking impacts on the banking system: prudential and profitability metrics
- Metrics computed: Basel III Leverage Ratio (LR), Liquidity Coverage Ratio (LCR), Net Stable Funding Ratio (NSFR), Net Interest Margin (NIM), Return on Equity (ROE). Box 2 provides computation details.
- Effects by scenario (Table 4: Prudential Ratios, NIM, ROE, and Loans in Each Scenario). Values shown exactly as in source:

  - Baseline Scenario values and scenario outcomes:
    - LR: Baseline 0.050; Scenario 0 0.050; Scenario 1 0.050; Scenario 2-A 0.050; Scenario 2-B 0.048; Scenario 3-A 0.050; Scenario 3-B 0.050; Scenario 4-A 0.051; Scenario 4-B 0.053
    - LCR: Baseline 1.40; Scenario 0 1.40; Scenario 1 1.39; Scenario 2-A 0.65; Scenario 2-B 0.57; Scenario 3-A 1.12; Scenario 3-B 0.90; Scenario 4-A 0.44; Scenario 4-B 0.70
    - NSFR: Baseline 1.09; Scenario 0 1.09; Scenario 1 1.09; Scenario 2-A 0.98; Scenario 2-B 0.92; Scenario 3-A 0.95; Scenario 3-B 0.92; Scenario 4-A 0.95; Scenario 4-B 0.99
    - NIM: Baseline 0.0278; Scenario 0 0.0278; Scenario 1 0.0275; Scenario 2-A 0.0264; Scenario 2-B 0.0248; Scenario 3-A 0.0243; Scenario 3-B 0.0256; Scenario 4-A 0.0273; Scenario 4-B 0.0256
    - ROE: Baseline 0.087; Scenario 0 0.087; Scenario 1 0.084; Scenario 2-A 0.053; Scenario 2-B 0.043; Scenario 3-A 0.009; Scenario 3-B 0.033; Scenario 4-A 0.055; Scenario 4-B 0.010
    - ΔLoans: Baseline (blank); Scenario 0 0; Scenario 1 0; Scenario 2-A 0; Scenario 2-B 0; Scenario 3-A 0; Scenario 3-B 0; Scenario 4-A -4%; Scenario 4-B -8%
  - SOURCE: IMF staff estimates.

- Key implications from metrics:
  - Prudential ratio impacts emerge primarily in scenarios where the central bank expands its balance sheet (Scenarios 2 and 3).
  - In Scenarios 2-A, 2-B, and 3-B, LCR and NSFR fall below regulatory minima due to increased short-term funding and the reduction in deposit funding (deposits treated as long-term in calculations).
  - Scenario 2-A sees an impact on LR as commercial banks expand balance sheets to obtain additional government bonds for collateral.
  - Profitability (ROE, NIM) comes under pressure across Scenarios 2 and 3 as banks replace cheap deposit funding with more expensive funding sources (central bank funding, wholesale funding, or both), while loan portfolios are initially unchanged.
  - Scenario 4 second-round effects lead to reductions in loans: Scenario 4-A (loans contract due to higher loan rates); Scenario 4-B (further loan contraction due to higher deposit remuneration and higher overall funding costs).
  - In Scenarios 0 and 1, there is no or virtually no impact on prudential and profitability metrics: a central bank liability is exchanged for another, with little effect on the banking system despite a drain in deposits in Scenario 1 because reduction is accommodated by ample reserves.

### Box 2 summary: prudential ratios and profitability indicators (computation notes)
- Metrics tracked: LR, LCR, NSFR, plus bank profitability measures.
- LR definition used: LR = High-quality capital / Total non-risk-weighted assets.
  - Baseline LR in analysis: 5 percent (0.050), just above Basel III minimum requirement of 3 percent.
- LCR definition and assumptions:
  - LCR = HQLA / Total net cash outflows over next 30 days.
  - Numerator: stock of HQLA; government bonds must be unencumbered and exclude sovereign bonds already used as collateral for central bank borrowing (assume zero haircuts; volume of collateral equals central bank lending amount).
  - 20 percent of Other Assets considered HQLA (includes high-rated corporate debt and mortgage-backed securities).
  - Denominator: assume 5 percent of deposits are flighty; 50 percent of other short-term liabilities are cash outflow items (liabilities < 30 days); existing debt assumed long-term and excluded from HQLA needs.
- NSFR and other details referenced as part of Basel III framework; scenario analysis limited to on-balance sheet exposure for simplicity.

### Key takeaways and considerations
- Two contrasting central bank outcomes:
  - CBDC issuance that can be absorbed by reducing other liabilities or excess reserves (Scenarios 0 and 1) leaves central bank and bank balance sheets largely unchanged.
  - CBDC issuance that requires additional reserves forces central bank asset expansion (Scenarios 2, 3, 4), with material implications for banking sector liquidity and profitability.
- Whether and how reserves are replenished (central bank lending vs. asset purchases vs. domestic/foreign wholesale funding) is central to the banking system’s adjustment path and prudential outcomes.
- Second-round effects (changes in deposit and lending rates) are the primary channel through which lending to the real economy contracts (Scenarios 4-A and 4-B).
- Preconditions matter: analysis assumes the bulk of payments are deposit-based and households hold no material cash; if households hold substantial cash, initial impacts on banks could be reduced.

*Source: IMF Working Paper — Central Bank Digital Currencies and Financial Stability: Balance Sheet Analysis and Policy Choices (Box 1 content).*

### Box 2. Prudential Ratios and Profitability Indicators (concluded)

### Box 2. Prudential Ratios and Profitability Indicators (concluded)

### The Net Stable Funding Ratio
- NSFR complements the LCR by more directly limiting banks’ reliance on potentially more volatile wholesale funding.
- Based on the Basel III definition,5 the NSFR is:
  - 푁푁푆푆 푁푁퐿퐿=
    퐻퐻퐺퐺푇푇푙푙푇푇푇푇푏푏푇푇푛푛 푇푇푎푎푇푇표표푛푛푇푇 푇푇표표 푐푐푇푇푇푇푏푏푇푇푛푛 표표표표푛푛푏푏푙푙푛푛푙푙
    퐿퐿푛푛푅푅 표표푙푙 푅푅푛푛푏푏 푇푇푎푎푇푇표표푛푛푇푇 푇푇표표 푐푐푇푇푇푇푏푏푇푇푛푛 표표표표푛푛푏푏푙푙푛푛푙푙
    =
    95%×퐷퐷푛푛퐷퐷 푇푇푐푐푙푙푇푇푐푐+30%×퐷퐷푛푛푏푏푇푇 푙푙푐푐푐푐표표푛푛푏푏+퐿퐿푇푇퐷퐷푙푙푇푇푇푇푇푇
    퐿퐿푇푇푇푇푛푛+30%×푂푂푇푇 ℎ푛푛푅푅 푇푇푐푐푐푐푛푛푇푇푐푐
- Assumptions used in the NSFR computation in the box:
  - 95 percent of retail deposits are assumed to be stable.
  - 30 percent of the debt issued has a maturity above 1 year.
  - Central bank lending and other short-term liabilities have a maturity below 6 months and are therefore not included in the computation of the NSFR.
  - On the asset side, all the loans are encumbered for a period of one or more years.
  - Government bonds are safe (and therefore do not appear in the denominator of the NSFR).
  - 30 percent of the other assets are assumed to be of lower quality with a long-term maturity.

### Profitability
- Profitability is not a central bank policy objective per se, but reductions in profit can have financial stability implications:
  - An increase in funding costs (e.g., from a rise in wholesale funding) can transmit to higher lending rates and reduce provision of credit, all else equal.
  - An unprofitable banking sector will see capital erode over time, creating incentives to take risks when capital becomes thin and leverage high.
- ROE is measured as:
  - 퐿퐿푂푂푅푅=(푁푁푁푁푁푁×푇푇푇푇푇푇푇푇푇푇 퐻퐻푐푐푐푐푛푛푇푇푐푐+푂푂푇푇 ℎ푛푛푅푅 퐿퐿푛푛퐺퐺푛푛푛푛표표푛푛푐푐−푂푂푇푇 ℎ푛푛푅푅 푅푅퐸퐸퐷퐷푛푛푛푛푐푐푛푐푐 )/퐿퐿푇푇퐷퐷푙푙푇푇푇푇푇푇
- NIM (net interest margin) is defined as:
  - 푁푁푁푁푁푁=
    (
    푁푁푛푛퐺퐺푛푛푇푇푎푎푛푛푛푛푇푇 푙푙푛푛푇푇푎푎푛푛   −푁푁푛푛푇푇푛푛푅푅푛푛푐푐푇푇 푛푛퐸퐸퐷퐷푛푛푛푛푐푐푛푛푐
    )/(푇푇푇푇푇푇푇푇푇푇 퐻퐻푐푐푛푛푇푇푐푐)
  - =(5%×퐿퐿푇푇푇푇푛푛+3%×퐺퐺퐶퐶 −0.1%×퐷퐷푛푛퐷퐷 푇푇푐푐푙푙푇푇푐  −2%×퐿퐿퐶퐶 푇푇푛푛푛푛푏푏푙푙  푛푛푙푙 −4%×퐷퐷푛푛푏푏푇푇 푙푙푐푐푐푐표표푛푛푏푏
    −2%×푂푂푇푇 ℎ푛푛푅푅 푆푆푇푇)/(푇푇푇푇푇푇푇푇푇푇 퐻퐻푐푐푛푛푇푇푐푐)
- Values of the various interest rates (on loans, deposits, CB lending...) are based on euro area and United States (US) data.
- Baseline scenario:
  - The values of Other Revenues and of Other Expenses are such that the ROE is around 9 percent.6
  - Footnote context: Based on Stiroh (2004), the value for Other Revenues is set at 185, representing 40 percent of the entire revenue. Other Expenses are assumed to be 420, implying an ROE of around 9 percent. For comparison, the ROEs of the euro area and the US banks have been around 5 and 10 percent since the GFC, respectively, according to ECB data and FRED.
- CBDC issuance effects on profitability:
  - Will weigh on the banking sector profitability through NIM, Other Revenues, and Other Expenses.
  - A drop in deposits will reduce services usually charged by banks on those deposits; Stiroh (2004) indicates charges related to these services represent roughly 5–6 percent of total revenues.
  - The analysis assumes that service charges accounting for 5 percent of entire revenues decline in proportion to the drop of deposits.7

### Potential Implications for Financial Stability (summary of main points)
- Main potential implications of CBDC introduction (section starts from scenario outcomes):
  1. A large substitution of CBDC for deposits can make it more difficult for banks to adhere to key prudential constraints, such as the LCR, NSFR, and the LR, unless excess reserves are ample.
  2. Increased competition for deposits and replacement of low-cost deposit funding with central bank- or wholesale funding reduces banks’ profits. Some banks may respond by increasing risks, exiting the market, or consolidating.
  3. Loss of deposit funding likely leads banks to increase the share of wholesale funding, strengthening interconnections with NBFIs and increasing exposure to funding shocks; reliance on foreign-currency funding would raise FX risk management needs.
  4. If banks replace deposits with central bank borrowing, demand for sovereign bonds may rise because central bank loans are typically collateralized with HQLA, potentially entrenching the sovereign-banks nexus.
  5. Banks may reduce credit provision to the private sector if funding costs increase, and structural changes may accelerate as customers move away from banks for savings.
  6. Financial stability risks are likely most pronounced during stress times: widely accessible CBDC seen as a “safe haven” can induce runs from individual banks or the banking system into CBDC, especially dangerous for weaker banks.

### Summary of Scenarios and Key Financial Stability Implications (as presented in Table form)
- Key driver: the size of the drain on deposits relative to total assets; severity increases with larger deposit outflows and depends on reserves, availability of other HQLA, other domestic funding sources, and competitive environment.
- Scenario matrix (headings preserved from source):
  - Scenario 0 Cash: CBDC replacement = Cash; Excess reserves = No; Reserve offset mechanism = Limited impacts; Second round effects = (blank); Financial stability implications = Limited impacts.
  - Scenario 1 Deposits: CBDC replacement = Deposits; Excess reserves = Sufficient; Reserve offset mechanism = No; Financial stability implications = Limited impacts.
  - Scenario 2-A Deposits: Excess reserves = Insufficient; Reserve offset mechanism = CB lending; Second round effects = No; Financial stability implications = 1. LCR, NSFR; 2. Profits decline.
  - Scenario 2-B Deposits: Excess reserves = Insufficient; Reserve offset mechanism = CB lending (balance sheet expansion); Second round effects = No; Financial stability implications = 1. LCR, NSFR, LR; 2. Profits decline; 3. NBFI funding; 4. SB Nexus.
  - Scenario 3-A Deposits: Excess reserves = Insufficient; Reserve offset mechanism = Domestic borrowing; Second round effects = No; Financial stability implications = 1. LCR, NSFR; 2. Profits decline; 3. NBFI funding; 4. SB Nexus.
  - Scenario 3-B Deposits: Excess reserves = Insufficient; Reserve offset mechanism = Foreign borrowing; Second round effects = No; Financial stability implications = 1. LCR, NSFR; 2. Profits decline; 3. NBFI funding; 4. SB Nexus.
  - Scenario 4-A Deposits: Excess reserves = Insufficient; Reserve offset mechanism = Loan shrinks, CB lending; Second round effects = Yes (lending rates); Financial stability implications = 1. LCR, NSFR; 2. Profits decline; 5. Credit contraction.
  - Scenario 4-B Deposits: Excess reserves = Insufficient; Reserve offset mechanism = Loan shrinks, CB lending; Second round effects = Yes (deposits and lending rates); Financial stability implications = 1. LCR, NSFR; 2. Profits decline; 5. Credit contraction.
- Note: SB stands for sovereign-bank.

### Breach of Prudential Ratios
- Several scenarios can cause banks to breach LCR, NSFR, and potentially LR when deposit outflows force banks to rely on short-term funding (central bank lending and wholesale funding).
- Mechanisms:
  - Deposits are considered stable funding with lower implicit risk weighting; replacement increases reliance on funding considered more risky, lowering LCR and NSFR.
  - Increased central bank borrowing requires pledging additional collateral, encumbering assets and reducing HQLA available for LCR calculation.
  - Replacing deposits with wholesale funding and purchasing government bonds used as collateral can reduce the LR (increase leverage), as in Scenario 2-B; banks close to LR minima can breach the requirement.
- Adjustment options (all typically reduce profitability or lending):
  - Raise longer-term funding (longer than 30 days for LCR; longer than at least 6 months and ideally 1 year for NSFR) — increases funding costs and reduces profitability.
  - Boost share of HQLA by reducing loans or risky investments — increases LCR/NSFR but weighs on profits and loans provided.
  - Raise remuneration of deposits to limit outflows — could maintain ratios but reduces profitability and may limit CBDC holdings in equilibrium.

### Reduction in Profits and Greater Banks’ Risk Taking
- Competitive reactions to CBDC may increase rates paid on remaining deposits, raising funding costs and lowering bank profits and franchise values.19
- Four channels through which CBDC can affect competition, profits, and lending:
  1. Existing level of competition matters:
     - In less competitive systems (monopoly rents), banks may absorb increased costs and reduce impact on loans.
     - In highly competitive markets (near-zero economic profits), banks pass costs to lending rates, leading to stronger reductions in credit provision.
  2. Risk-shifting:
     - Higher funding costs create incentives for increased risk taking (risk-shifting), stronger for more leveraged banks and weaker for better capitalized banks.
     - Incentives to take risk are smaller the larger the remaining franchise value shareholders would forfeit by increasing risk.20,21
     - Risk-shifting may arise cross-sectionally for least well-capitalized and least profitable banks and may strengthen over time as profitability erodes capital.
  3. Informational frictions:
     - Deposit outflows can reduce banks’ access to customer information, worsening credit decision-making, increasing credit risk, and reducing credit supply.
     - Offsetting effect: if banks distribute CBDC or participate on the payment platform, they may retain or gain information on transactions; platform design could even increase information availability to competing banks, affecting competition (see Box 3 in source).
  4. Exit and consolidation:
     - Reduced profitability may lead some banks to exit or be taken over, reducing the number of active banks (Chiu et al. (2023)).
     - Consolidation may be desirable in “overbanked” markets but can create larger banks that may become systemically important (too-big-to-fail), requiring sufficient capital and tighter regulation.22,23
     - The FSB evaluation notes remaining gaps in resolution reforms and that reforms have shifted some activities to nonbanks.24

*IMF Working Paper — Central Bank Digital Currencies and Financial Stability: Balance Sheet Analysis and Policy Choices — Box 2 (concluded).*

### Box 3. Impact on Financial Institutions’ Access to Information on Borrowers’

### Box 3. Impact on Financial Institutions’ Access to Information on Borrowers’ Credit Quality Depending on Design of CBDC

### Role of payment data for credit assessment
- In the absence of a CBDC, clients’ payment activities are visible to banks and generate roughly 90 percent of banks’ useful customer data (McKinsey (2019)).
- Empirical studies documenting the use of payment data for credit decisions include Puri et al. (2017) (German consumer data), Mester et al. (2007) (Canadian small business data), and Hau et al. (2019) (Ant Financial loans to online vendors).
- Disruptions from fintech payment providers reduce banks’ synergies between payment and credit and can weigh on bank lending (Parlour et al. (2022); Ghosh et al. (2021)).

### How CBDC design affects banks’ access to payment information
- The effect of CBDC on banks’ ability to assess borrower credit quality depends heavily on CBDC design.
- If banks participate in CBDC distribution, they may retain information on payments made by their clients using the CBDC.
- If all participating banks on a CBDC payment platform are given access to information on all CBDC transactions made on it, available information could increase substantially.
  - In that case, proprietary information formerly held by individual banks becomes available to competitors, likely increasing competition in credit markets.
- There is a trade-off between:
  - Ensuring credit quality information is available to the banking sector, and
  - Customers’ preference for privacy and the desire for some anonymous payments (see ECB (2022) cited).

### Consequences of deposit falls and shifts in funding
- A fall in deposits following CBDC introduction (in all scenarios other than Scenario 0) could alter banks’ informational access with implications for credit markets.
- Issuance of CBDC can lead to growth in the share of wholesale funding—typically sourced from NBFIs—in several scenarios.
  - Increased reliance on short-term wholesale funding was a key element in the GFC in 2007–09 and resurfaced in the COVID-19 pandemic context.
- Domestic NBFIs holding safe assets at CBDC issuance may receive liquidity inflows and ultimately lend to banks on the wholesale market (illustrated in Scenarios 2-B and 3-A).
  - NBFIs would rebalance portfolios towards the banking sector: commercial paper, longer-term bank unsecured debt, or collateralized bank debt (e.g., covered bonds).
- Where domestic alternatives to deposit funding are scarce, banks could tap cross-border funding from foreign NBFIs, likely in foreign currency (Scenario 3-B).
- If banks increase reliance on NBFI funding, the financial system could become more exposed to funding shocks and adverse effects on real-economy financing.
  - A shock triggering large redemptions from open-ended investment funds could dry up funding for NFCs and banks.
  - Large market moves and margin calls could force NBFI investors to liquidate positions, reducing their ability to fund banks or NFCs.
  - CBDC could add to the flightiness of wholesale funding if households redeem fund shares in favor of CBDC safety.
- Increased reliance on foreign NBFIs raises FX and liquidity risks; difficulties rolling over FX funding can cause liquidity problems in derivatives markets (March 2020 “dash-for-cash” episode cited).

### Effects on bank funding costs, lending volatility, and credit provision
- Substitution of deposits by wholesale funding could increase volatility of banks’ funding costs and lending rates.
  - Banks funded more heavily with core deposits provide more loan-rate smoothing; loss of these deposits reduces that cushioning (Berlin and Mester (1999)).
- Two main channels can reduce credit supply to the real economy:
  1. Loss of deposit relationships reduces information and foreclosure/enforcement power (e.g., banker’s setoff), increasing lending risk and lending rates.
  2. Competition for deposits forces banks to offer higher deposit rates or replace deposits with costlier wholesale funding, raising banking system funding costs that are passed to lending rates.
- Pass-through of higher funding costs to loan rates depends on market structure:
  - Fully passed through in competitive banking systems.
  - Partially absorbed if banks are monopolistic and lower profits instead.
- Larger banks may substitute deposits with wholesale funding more easily than smaller banks, advantaging larger banks (Whited et al. (2023)).
- Degree of financial inclusion and potential deposit inflows from CBDC adoption could offset some adverse effects (see Box 4 referenced).
- Adverse effects on credit provision are likely limited in scenarios where the central bank balance sheet is unchanged (e.g., CBDC exchanged one for one with cash or existing excess reserves).
  - By contrast, when CBDC issuance is against deposits and the central bank generates additional reserves for banks to purchase CBDC on behalf of customers, the central bank balance sheet expands, increasing “outside” money (liabilities of the central bank: physical and digital currency, and reserves), potentially to the detriment of “inside money” (bank deposits).

### Sovereign–bank nexus and collateral demand
- If banks replace deposit funding with central bank funding (typically collateralized), demand for high quality collateral rises, increasing banks’ holdings of sovereign bonds.
  - Sovereign bonds’ zero-risk weight treatment leaves risk-weighted assets unchanged but strengthens sovereign–bank links.
  - Strengthened nexus can create a doom loop between fiscal and financial sectors in shocks, especially in emerging markets with high sovereign indebtedness.
- Higher allocation of savings toward government assets and away from private-sector loans could induce higher government indebtedness and negative feedback loops between macro shocks, fiscal deterioration, and financial exposures to the sovereign.
- Increased demand for high-quality collateral may prompt search-for-yield behavior and increase systemic risk.

### Potential partial offsets and non-bank credit provision
- Some reduction in bank credit could be compensated by increased credit provision by non-banks, particularly if nonbanks distribute the CBDC and gain access to payment information.
- However, NBFIs do not automatically gain additional funding sources when households hold CBDC (households increase holdings of assets with the central bank rather than the NBFI sector).

### Empirical and cross-country evidence
- Aggregate evidence: a clear negative correlation exists between the ratio of cash to deposits and the amount of credit to the private sector as a share of GDP (Figure 1), across a sample of 131 advanced and emerging market and developing economies.
  - Note: the correlation may not be causal.
  - The ratio in Figure 1 increases from near zero to near unity across countries in the sample; CBDC introduction alone is unlikely to produce such large cash-to-deposit ratio increases.
- Selected studies:
  - Merrouche and Nier (2012): payment system reforms (1995–2005) associated with trend decrease in use of currency relative to demand deposits and trend increase in credit to the private sector via larger funds intermediated by banks.
  - Grodecka-Messi and Zhang (2023): granting monopoly in issuance of banknotes to Bank of Canada (1935) reduced other banks’ profits but had no effects on lending, perhaps because deposit funding remained available.
  - Peek and Rosengren (2000), Khwaja and Mian (2008): liquidity pressures on banks can reduce credit provision in times of stress.
  - Forbes et al. (2022): banks with higher shares of NBFI or US dollar funding experienced higher stress at COVID-19 onset (measure of stress: change in banks’ CDS spread).
- Cross-country factors affecting cash use include credit card industry structure, average card transaction value, and merchant acceptance (Bagnall et al. (2014) referenced).

### Steady state, adoption path, and crisis considerations
- Adjustment to a new steady state is likely gradual:
  - Possible path: higher share of non-deposit funding or higher-remunerated deposits → more volatile/costly funding passed to higher steady-state loan-rates → lower provision of bank credit in steady state.
- Operational challenges can arise during adoption even if initial uptake is slow; behavioral changes could lead to rapid CBDC take-up creating financial stability challenges.
- In crisis times, households may switch into CBDC as a flight to safety, altering behavior even after steady state.

*Source: IMF Working Paper — Box 3. Impact on Financial Institutions’ Access to Information on Borrowers’ Credit Quality Depending on Design of CBDC*

### Box 4. Approaches to Modelling the Effect of a CBDC on Bank Lending

### Box 4. Approaches to Modelling the Effect of a CBDC on Bank Lending

### Modelling approaches and core mechanism
- Studies differ in assumptions regarding bank competition, funding sources, and whether the CBDC can replace cash or a range of deposits.
- Standard theoretical result: when funding costs increase, there will be a pass-through to loan rates, which will reduce credit as loan demand falls. The strength of the effect depends on the degree of competition in loan markets.
  - Imperfect competition: only partial pass-through to loan rates and a more muted reduction in lending.
  - More competitive markets: increases in funding costs are fully passed on, producing a more pronounced reduction in loan volumes.
- Modelling choices that drive divergent outcomes include:
  - Whether wholesale funding can replace deposit funding.
  - CBDC design features, notably whether CBDCs are remunerated or not.
  - Channels through which households substitute among cash, time deposits, demand deposits, and CBDC.

### Examples from the literature (model setups and outcomes)
- Keister and Sanches (2023)
  - Banking sector: perfectly competitive; banks make zero profits; wholesale funding not available.
  - Constraint: asset pledgeability constraint.
  - Result: introducing an interest-bearing CBDC raises banks’ funding costs and reduces credit creation, causing some disintermediation. The CBDC can still be welfare enhancing when households value improved payment ability.
- Whited et al. (2023)
  - Model: links between deposit and lending markets under imperfect competition; CBDC introduced counterfactually as a bundle of existing payment characteristics.
  - Demand: consumers have demand for CBDC even when it is not remunerated.
  - Quantitative findings:
    - A non-interest bearing CBDC would capture around 8 percent of the deposit market.
    - Leads to just over a one-percentage fall in lending compared to a banking industry without a CBDC.
    - Banks’ profits are reduced, weighing on banks’ capital and lending capacity.
  - Mitigating factor: banks’ ability to replace deposits with wholesale funding mitigates disintermediation, but wholesale funding is more interest sensitive than deposit funding, increasing interest rate risk exposure.
  - Distributional effect: small banks, with less access to wholesale funding, have more difficulty adapting than large banks.
  - When the CBDC is remunerated:
    - Banks would respond by raising deposit rates.
    - Banks would still lose up to 30 percent of their deposits if the CBDC paid the federal funds rate.
    - Lending would decline by more than 7 percent compared to a non-CBDC case.
- Andolfatto (2020), Chiu et al. (2023), Chang et al. (2023)
  - Setup: monopolistic or similar banking market structures.
  - Finding: issuance of a remunerated CBDC can improve bank lending via higher deposit rates that incentivize depositors to save more and induce previously un-banked to switch from cash into deposits; CBDC-induced higher remuneration on deposits can reduce cash holdings and increase deposits.

### Risks in the context of adoption
- Initial adoption volumes may be low due to network effects and behavioral inertia favoring existing payment systems.
- Network externalities can give rise to multiple equilibria: both “no one joining” and “everyone joining” are Nash outcomes.
- A tipping point is conceivable where incentives trigger a flip to broad CBDC adoption, potentially triggering disruptive adjustments (for example, banks curtailing credit).
- Behavioral substitution effects:
  - Households using CBDC instead of deposit-based payments may reduce the value of maintaining bank relationships, prompting transfers from checking and savings accounts into brokerage accounts with non-banks.
  - Businesses: initially receive CBDC from retail customers but may continue to pay suppliers and employees via deposit-based systems; widespread adoption could prompt payroll and supplier payments in CBDC, increasing business-to-business CBDC use and overall CBDC payment volumes.

### Risks in periods of stress
- Existence of CBDC alongside bank deposits may facilitate run-like behavior when doubts arise about solvency of individual banks or banking-system stability.
- Digitalization accelerates runs: social media and digital banking make bank runs faster than in the past (e.g., US regional banking turmoil of Spring 2023).
- Mechanism: households holding both deposit accounts and CBDC can rapidly switch deposits into CBDC, enabling quick outflows from a stricken bank and increasing probability and speed of runs, potentially forcing fire-sale liquidation of credit portfolios.
- System-wide runs:
  - With CBDC, a domestic alternative to physical cash or foreign deposit accounts exists, making exchange of risky banking claims for central-bank liabilities easier and potentially increasing system-wide run probability at lower perceived system risk.
  - System-wide runs may be especially concerning for reserves currency issuers or central banks with high credibility (leading to large inflows into the central-bank liability).
  - In jurisdictions with weaker central-bank credibility, flight to safety may more likely continue into reserves currencies (e.g., the dollar).
- Possible emergency responses:
  - System-wide “controls” on conversion of deposits into CBDC, akin to capital outflow controls used in crises.
  - At the individual-bank level, banks could refuse customer demands to purchase CBDC, analogous to closing doors during traditional runs.
- Heterogeneous vulnerability:
  - Banks most exposed to the aggregate shock or with high shares of uninsured deposits are likely to suffer larger deposit outflows to CBDC.
  - Runs on weakest institutions are likely to emerge first; containment determines whether they trigger system-wide runs.
- Liquidity and lender-of-last-resort considerations:
  - Weaker banks may need heavy central-bank borrowing using lenders-of-last-resort facilities unless CBDC design reduces run potential.
  - CBDC issuance could exacerbate liquidity problems in the NBFI sector during stress unless central banks revisit lists of eligible counterparties.

*Source: Box 4. Approaches to Modelling the Effect of a CBDC on Bank Lending (from the supplied IMF Working Paper content)*

### 5.     Policy Options to Mitigate Adverse Effects of

### 5.     Policy Options to Mitigate Adverse Effects of CBDCs on Financial Stability

### Overview
- Main policy levers considered: macroprudential policy, monetary policy and central bank market operations, CBDC design, and private digital payment arrangements using central bank settlement.
- Key conclusion: mitigating emerging risks by macroprudential or regulatory policies alone is likely to be difficult; monetary policy and central bank market operations may not fully neutralize CBDC effects; expansion of central bank lending can put central bank balance sheets at risk.
- Primary mitigation is likely to come from CBDC design that encourages CBDC use as a means of payment (extensive margin) while inducing households and firms to hold small amounts for retail payments (intensive margin).
- Design tools highlighted: holding caps, zero or negative remuneration, and interoperability with the banking system. Using CBDC to promote payment systems based on bank deposits can preserve payment efficiency and contain financial-stability costs.

### Macroprudential Policy
- Introducing a CBDC can reduce prudential ratios: both the LCR and the NSFR are expected to fall with the drop in deposits; in some scenarios (Scenario 2-B) leverage can increase and the LR can fall below its regulatory minimum.
- Relaxing prudential requirements after CBDC introduction is unlikely to be justified from a financial stability perspective because the drop in deposits increases funding risk.
- Policy responses that may be warranted:
  - Hike banks’ capital buffers ahead of CBDC launch to reduce incentives for risk-taking, strengthen the bank-sovereign nexus, and reduce run likelihood.
  - Introduce caps on FX funding or LCR ratios applied separately for each major funding currency where banks increase reliance on FX funding.
  - Introduce stressed debt service-to-income ratios to address higher financial instability from more volatile lending rates.
- If CBDC issuance increases banks’ funding from domestic or foreign NBFIs, progress in macroprudential regulation of the NBFI sector is needed (e.g., strengthened regulation of money market funds and open-ended investment funds).
- Limitations:
  - Macroprudential tools work imperfectly and cannot fully counter drops in credit supply due to higher funding costs or frequent depositor runs.
  - Adjustments to other central bank policies and CBDC design options may ultimately hold more promise.

### Monetary Policy and Central Bank Market Operations
- Monetary Policy
  - Central bank can take changes in financial conditions into account when setting policy rate to achieve price and output stability objectives.
  - Central bank could tighten or loosen policy in response to CBDC-induced financial conditions to partially offset yield curve shifts.
  - Challenge: interest rate impacts may be non-uniform across markets (deposit rates may rise while government bond yields may fall), making a single policy-rate response inadequate.
  - Policy rate is not the right tool to address doubts about banking-system strength or relative shifts between deposits and CBDC in a crisis.

- Central Bank Market Operations
  - Expanding central bank lending via open market operations can generate additional reserves for the banking system to purchase CBDC, but three limits mean such lending may not neutralize detrimental effects:
    1. Central bank funding is typically more expensive than deposit funding (deposit rates near zero vs central bank lending tied to positive policy rates), raising banks’ funding costs and reducing profits and lending.
    2. Central banks typically accept only high-quality liquid assets (HQLA) as collateral; expanded lending against HQLA does not alleviate pressure on banks’ LCR if unencumbered HQLA falls. Broadening collateral to include loan pools can help but transfers credit risk to the central bank and may be unattractive, especially in EMDEs.
    3. Central bank lending is typically short-term (day or week); short-term central bank funding does not improve banks’ NSFR because it does not count as long-term stable funding even if rolled over.
  - Extending lending to longer tenors (beyond one year) could improve NSFR, and central banks have used LTROs/TLTROs in acute liquidity stress. However, offering long-term lending in steady state effectively substitutes central banks for the banking system in maturity transformation and is unattractive outside crises.

- Lender of Last Resort (LOLR)
  - In a “digital bank run” where deposits shift rapidly into CBDC, the central bank can act as LOLR to provide emergency funding and prevent bank collapses by lending CBDC proceeds back to banks.
  - Practical challenges:
    - Traditional LOLR lends only to solvent banks and against collateral to avoid central bank losses. Fragile banks may be most exposed to deposit outflows, reducing scope for standard LOLR.
    - Banks’ assets are often illiquid, blurring illiquidity vs insolvency during stress.
  - Possible enhancements:
    - Forward-looking solvency assessments (e.g., credible prospect to regain solvency within 6 months) could allow temporary support for undercapitalized banks or use of government guarantees.
    - Ex ante expansion of acceptable collateral via system-wide valuation and prepositioning of assets, with committed haircuts, to enable advances against broader asset classes.
    - Central bank monitoring as operator of the CBDC payment system can help identify affected banks and target timely liquidity injections.
  - Trade-off: operational and balance-sheet risks to the central bank increase when expanding LOLR capacity; such actions may be appropriate in crises but not attractive in steady state.

### Central Bank Design of CBDC
- Objective: design CBDC to be used primarily as a means of exchange (payments) rather than a store of value, to avoid crowding out deposits while delivering policy objectives.
- Central banks face a trade-off: sufficient adoption to achieve objectives vs excessive adoption that crowds out bank deposits and triggers adverse scenarios.
- Three interrelated design elements considered: holding caps, remuneration (zero/negative), and interoperability with the banking system.

- Caps on Holdings of CBDC
  - Aggregate issuance cap: easy to implement (stop issuance when desired amount reached) but differs from cash (which is always provided on demand). Downsides:
    - Early hoarding by some individuals; others unable to acquire CBDC as cap is reached.
    - In crises, a secondary market could emerge with CBDC trading above par.
    - Aggregate limits may incentivize pre-emptive switches to CBDC.
  - Per-person caps via wallet limits: wallets (digital or hardware) can limit holdings; excess could auto-convert to a bank deposit account (example noted: Bahamas Sand Dollar).
  - Circumvention risk: multiple wallets per person; mandatory identification/digital identity can prevent circumvention but trades off anonymity/privacy goals.
  - Transaction-size and frequency caps: easier to enforce (built into payment system); may be circumvented by algorithmic splitting but a combination of amount and frequency caps can effectively limit retail use.
  - Temporary “kill switch”: central bank could stop convertibility into CBDC to break a run; this is an aggregate cap and may be ineffective when runs target individual banks.

- Remuneration of CBDC
  - Current practice: all CBDCs in circulation or planned have zero interest; motivations include mimicking cash, technical/legal constraints, and reducing attractiveness as a store of value.
  - Positive remuneration would increase adoption but exacerbate competition with deposits and credit disintermediation risks.
  - Alternatives:
    - Tiered wallet structure with different interest rates by holding size; negative interest rates on holdings above a threshold to make large holdings costly.
    - Negative rates can be more market-friendly than strict caps but may be politically unpopular, seen as a tax on money, and undermine CBDC support.
    - To limit competition with deposits, interest on CBDC would generally be at most zero in many designs.

- Interoperability with the Banking System
  - Convertibility between CBDC and bank deposits should be ensured similar to deposit-to-cash convertibility.
  - Banks can play distribution roles for CBDC; if CBDC wallets are offered by banks, banks can retain customer relationships and continue to offer services even when deposits are converted to CBDC.
  - Interoperability designs:
    - Wallets interoperable with bank accounts so funds flow instantly; prepaid volumes with automatic transfers from bank accounts when exceeded.
    - On-demand creation and retirement of CBDC at payment time (CBDC created when needed, retired when received) to minimize outstanding CBDC balances.
  - Benefits: reduces CBDC as a store of value, lowers adverse effects on credit intermediation, maintains banks’ role in payments.
  - Trade-offs: loss of transaction data to banks if CBDC is not intermediary-mediated; privacy-preserving designs may limit banks’ access to payment histories and affect credit assessment.

- Wholesale CBDC and Efficiency of Retail Payments
  - Wholesale CBDC (for financial institutions only) can achieve policy goals like enhancing retail payment efficiency by facilitating interoperability between tokenized deposits.
  - Wholesale CBDC can provide settlement in central bank money for net interbank claims from tokenized-deposit platforms, preserving singleness of money.
  - Example: tokenized deposits initiatives (e.g., USDF) produce net interbank positions that require settlement in central bank money (e.g., Fedwire in absence of wholesale CBDC).
  - CBDC designs enabling private deposit-based payment solutions (tokenized deposits or fast payments) have no direct effect on credit intermediation and can preserve payment efficiency while containing retail CBDC risks.
  - Central bank choice: make CBDC widely available (extensive margin) while discouraging large holdings (intensive margin) via remuneration, caps, and interoperability.
  - Empirical design indications noted: Bank of England proposes limits between £10,000 and £20,000 per individual; ECB considers around 3,000 to 4,000 digital euros per capita.

### Other Financial Sector Regulatory and Banking Sector Policies
- Where banking sectors are fragile, CBDC introduction could be particularly risky; authorities should:
  - Strengthen the banking sector before launch by addressing structural weaknesses, requiring higher capital, and ensuring solid supervision and regulation.
  - Close gaps in supervision, resolution regimes, and data availability for “too important to fail” institutions before widespread CBDC launch expected to induce exits and consolidation.
  - Maintain comprehensive supervision and regulation of the banking sector to reduce vulnerability exposure to deposit shifts to CBDC.
- Deposit insurance:
  - A strong and credible deposit insurance framework (where appropriate preconditions exist) can reduce incentive for depositors to switch to CBDC in crisis periods.
  - Coverage limits should be sufficiently large and compensation processes fast and reliable to deter runs.
- Cyber resilience and fraud prevention:
  - Digitalization increases cyber and fraud risks; efforts should be undertaken to increase system resilience, including maintaining physical cash as an alternative payment means to ensure robustness against outages.  

*IMF Working Paper — Chapter 5: Policy Options to Mitigate Adverse Effects of CBDCs on Financial Stability*

### 6.     Conclusion

### 6.     Conclusion

### Financial stability implications: overview
- The financial stability implications of the issuance of a retail CBDC can be mild, or material, depending on:
  - the design of the CBDC,
  - the resulting volume of CBDC,
  - the central bank’s strategy for issuing it,
  - pre-existing conditions shaping banking-sector competitive reactions.
- Adverse effects on the banking system become larger when the replacement of deposits is greater.
- If CBDC issuance leaves the size of the central bank balance sheet unchanged—as issuance is met one for one by a drawdown of physical cash held by households or excess reserves held by the commercial banking system—adverse effects on financial stability are unlikely to be material, even if the CBDC replaces deposits.
- Some effects can still arise if a loss in deposits leads to a loss in borrower information.

### Scenarios with central bank balance sheet expansion
- When issuance of the CBDC requires the central bank to generate a prior increase in reserves so commercial banks can buy the CBDC, the central bank’s balance sheet expands.
- In these scenarios adverse effects on financial stability are more likely via:
  - an increase in funding costs for banks,
  - a more volatile funding structure for banks,
  - ramped-up borrowing from NBFIs, the central bank, or both.
- In some scenarios the sovereign-bank nexus may be strengthened because more high-quality collateral is needed to accommodate greater borrowing from the central bank.
- Competitive bank reactions can exacerbate effects:
  - raising deposit remuneration increases funding costs on remaining deposits,
  - pressure on profitability can lead banks to take greater risks or exit, especially when capitalization is thin,
  - higher cost and volatility of bank funding can lead to crowding out of credit to the real economy (banks pass on higher funding costs and charge higher loan rates).
- Such adverse effects are likely only when the CBDC leads to an expansion of the central bank balance sheet; if the CBDC instead leads mostly to a reduction in cash or reserves, there is no such pressure on bank funding costs or loan rates.

### Dynamics in steady state and crisis
- Adverse effects can arise from:
  - changes in the structure of the financial system in steady state when CBDC issuance replaces deposits and is large enough to require prior creation of additional reserves,
  - rapid changes from multiple adoption equilibria,
  - crisis times, when doubts about solvency can lead to runs away from deposits and into the CBDC.

### Limits of regulatory and macroprudential responses
- Mitigating emerging risks through macroprudential or financial sector regulatory policies alone is likely to be insufficient.
  - Macroprudential policy can increase banks’ resilience but cannot address the risk of liquidity pressure or disruption of banks’ credit intermediation by CBDC.
  - Financial stability concerns intensify when CBDC yields a larger role of market-based finance and NBFIs in banks’ funding.
  - Given the diversity of NBFIs across actors and jurisdictions, and their international nature, achieving substantive progress in macroprudential control of NBFIs is challenging.

### Role of central bank lending operations
- Central bank lending operations can help reduce detrimental financial stability impacts in principle, both in steady state and in crisis times, if the central bank provides emergency liquidity to banks facing runs into the CBDC.
- Expanding lending operations against a broader set of collateral and at longer loan tenors would require central banks to be comfortable with possible losses and managing problem loans.
- Making such changes to lending operations in steady state is typically not palatable to central banks.
- Useful considerations include greater use of ex ante positioning of illiquid assets and assigning haircuts on these assets for emergency operations.

### CBDC design features to alleviate financial stability concerns
- Design features can foster CBDC use as a means of payment while discouraging large holdings that would hamper credit intermediation or shift banks toward wholesale funding.
- When remuneration of the CBDC is low—at zero or potentially negative for larger volumes held—limits to holding CBDCs and possible interoperability with the banking system would:
  - guarantee CBDC usability for transactions,
  - discourage its use as a store of value,
  - reduce adverse financial stability effects as CBDC adoption advances.
- Designing the CBDC for “wholesale” use among operators of deposit-based payment systems (including those based on tokenized deposits) can largely avoid adverse impacts on the financial system.
- A well-designed CBDC could co-exist with private payment alternatives and mitigate some financial stability concerns associated with securities-backed stablecoins.

### Key scenario-specific figures and illustrative calibrations (selected)
- Baseline consolidated banking sector balance sheet size: 10,000 (arbitrary unit).
  - Loans: 6,000 (60 percent of total assets).
  - Government bonds: 1,500 (15 percent).
  - Reserves: 140 (1.4 percent).
  - Others: 2,360 (23.6 percent).
  - Deposits (liabilities): 7,000 (70 percent of total liabilities).
  - Lending from the central bank: 750 (7.5 percent).
  - Debt issued: 500 (5 percent).
  - Capital: 500 (5 percent).
- Central bank balance sheet equals one fifth of banking sector: 2,000.
  - Central bank assets: securities 1,000 (half), lending to commercial banks 750 (37.5 percent), others 250 (12.5 percent).
  - Central bank liabilities: cash 1,000 (half), commercial bank reserves 140 (7 percent), others 860 (43 percent).
- Reserve requirements: 2 percent of deposits (example; footnote notes current rates: euro area 1 percent, US zero percent).
- Scenario 0 (cash-like CBDC):
  - CBDC issuance assumed at 400, representing 5 percent of the value of money (total deposits and cash used by households).
  - Result: 400 of cash replaced by 400 of CBDC on the central bank balance sheet; deposits unaffected.
- Scenario 1 (CBDC replaces deposits, banks possess excess reserves):
  - Initial reserves assumed to be 600; required reserves at 2 percent of deposits amount to 148 initially.
  - CBDC issuance: 400.
  - Excess reserves absorb full amount of CBDC; LCR and NSFR virtually unchanged.
- Scenario 2-A (CBDC replaces deposits, no excess reserves, banks possess enough collateral):
  - CBDC replaces deposits by 750 (around 9 percent share in households’ liquid assets).
  - Banks increase borrowing from the central bank by 750; central bank lending assumed short term.
  - LCR and NSFR decrease below regulatory minima of 100 percent.
- Scenario 2-B (greater CBDC replacement; commercial bank balance sheets expand):
  - CBDC replaces 1,200 of deposits (representing 15 percent of households’ liquid assets).
  - Banks lack sufficient government bonds; increase wholesale funding from NBFIs, purchase government bonds, use them to borrow from the central bank.
  - Commercial banks’ balance sheets expand; LCR and NSFR fall below regulatory minima of 100 percent; profitability falls.
- Scenario 3-A (banks’ balance sheet unchanged, higher wholesale funding):
  - Banks compensate deposit reduction by half short-term wholesale funding and half long-term funding (issued debt remunerated at 4 percent, twice the cost of central bank lending).
  - Resulting LCR: 1.12 (above 100 percent).
  - Resulting NSFR: 0.95 (slightly below 100 percent).
  - Profitability worsens significantly; ROE drops (precise ROE figure beyond excerpt).

*IMF Working Paper — Central Bank Digital Currencies and Financial Stability: Balance Sheet Analysis and Policy Choices*

### 0.9 percent.

### wpiea2024226-print-pdf - 0.9 percent.

### Scenario 3-B: Foreign borrowing
- Context and assumptions:
  - Domestic NBFIs are not willing or able to provide wholesale funding to the banking sector or sell their government bond holding to the central bank.
  - Banks rely on foreign borrowing exclusively in the form of short-term funding.
  - Lenders could be foreign NBFIs or foreign banks; foreign financial institutions are not willing to provide long-term lending in local currency.
  - Funding received from foreign institutions is short-term and denominated in a major foreign currency, such as the dollar or euro.
  - Compared to Scenario 3-A, a currency mismatch arises in the banking sector in addition to a maturity mismatch.
  - Central bank expands its balance sheet and increases its holdings of foreign government bonds; it purchases foreign currency from a domestic bank, which receives domestic currency reserves used to satisfy the move from deposits to CBDC.
  - The domestic bank borrows the foreign currency from a large international bank and increases its short-term wholesale funding in foreign currency.

- Balance sheet (Table 7) — consolidated banks and central bank (values preserved):
  - Consolidated banks — Assets:
    - Loans 6,000
    - Cash 1,000
    - GB (gov. bond) 1,500
    - Securities 2,200 (+1,200)
    - Reserves 140 (Reserves (require 116))
    - Others 250
    - Total 10,000
  - Consolidated banks — Liabilities:
    - Deposits 5,800 (-1,200)
    - Lending 750
    - CB lending 750
    - Debt issued 1,100 (+600)
    - Others 860
    - Other ST 1,850 (+600)
    - CBDC 1,200 (+1,200)
    - Capital 500
    - Total 10,000
  - Central bank — Assets:
    - Lending 750
    - Securities 2,200
    - Reserves 140
    - Others 2,360
    - Total 3,200
  - Central bank — Liabilities:
    - Reserves (require 116) 140
    - Debt issued 1,100 (+600)
    - Other ST 1,850 (+600)
    - CBDC 1,200 (+1,200)
    - Capital 500
    - Total 3,200

- Key risks highlighted:
  - Currency mismatch and maturity mismatch in the banking sector from short-term foreign currency funding.
  - Central bank balance sheet expansion and foreign asset accumulation.

### Scenario 4-A: Loans shrink
- Context and mechanism:
  - Central bank targets a relatively large CBDC issue size of 1,200 (as in Scenario 3).
  - Initial transactions generate reserves: loan from central bank to commercial banks collateralized by government securities, and outright purchase of government securities by central bank resulting in banks receiving wholesale funding from NBFI sector.
  - High funding costs on parts of commercial banks’ balance sheets transmit into higher loan rates; banks fully pass through the increase in funding costs to the loan rate, which increases by 0.3 percent.
  - Higher loan rates reduce loan demand; loan supply meets loan demand at a reduced volume, causing credit provision to shrink and allowing banking system to retire some expensive wholesale funding.
  - Banks rely on central bank financing (+960) to compensate for deposit drop (-1,200), but substitution is partial so banking sector balance sheet shrinks.

- Balance sheet (Table 8) — consolidated banks and central bank (values preserved):
  - Consolidated banks — Assets:
    - Loans 6,000
    - Cash 1,000
    - GB (gov. bond) 1,500
    - Securities 2,200 (+1,200)
    - Reserves 140 (Reserves (require 116))
    - Others 250
    - Total 10,000
  - Consolidated banks — Liabilities:
    - Deposits 5,800 (-1,200)
    - Lending 750
    - CB lending 750
    - Debt issued 500
    - Others 860
    - Other ST 2,450 (+1,200)
    - CBDC 1,200 (+1,200)
    - Capital 500
    - Total 10,000
  - Central bank — Assets:
    - CB lending 750
    - CB lending 750 (duplicate line per table structure)
    - Reserves 140
    - Total 3,200
  - Central bank — Liabilities:
    - Reserves (require 116) 140
    - Debt issued 500
    - Other ST 2,450 (+1,200)
    - CBDC 1,200 (+1,200)
    - Capital 500
    - Total 3,200

- Financial stability indicators and outcomes:
  - LR increases due to reduction in size of banks’ balance sheets.
  - LCR declines below regulatory minimum due to reduction in deposits and increase in central bank lending (decrease in unencumbered HQLA).
  - NSFR decreases as reliance on unstable central bank lending expands.
  - ROE decreases partly because the size of the balance sheet, including loans, is reduced.

### Scenario 4-B: Loans shrink together with the change in interest rates
- Context and assumptions:
  - CBDC issuance is 1,200, same as Scenario 4-A.
  - Banks replace some lost deposit funding with central bank funding; central bank purchases securities from NBFI sector.
  - Banks increase the deposit rate by 0.5 percent to prevent deposit outflow, but cannot prevent issuance of 1,200 of the CBDC.
  - Increase in funding costs from higher deposit rates and greater share of expensive central bank funding reduces bank profits and pushes banks to raise lending rates.
  - Compared to Scenario 4-A (full pass-through to lending), banks here pass through 75 percent of the increase in funding costs to lending rates.
  - Lending decreases by 40 percent of the CBDC issued (quantification referenced from previous literature).

- Balance sheets (Table 9) — two quantified cases preserved:
  - Case A (partial loan reduction shown first):
    - Consolidated banks — Assets:
      - Loans 5,760 (-240)
      - Cash 1,000
      - GB (gov. bond) 1,500
      - Securities 1,240 (+240)
      - Reserves 140 (Reserves (require 116) 140)
      - Others 250
      - Total 9,760
    - Consolidated banks — Liabilities:
      - Deposits 5,800 (-1,200)
      - Lending 1,710 (+960)
      - CB lending 1,710 (+960)
      - Debt issued 500
      - Others 860
      - Other ST 1,250
      - CBDC 1,200 (+1,200)
      - Capital 500
      - Total 9,760
    - Central bank — Assets:
      - CB lending 1,710 (+960)
      - Reserves 140
      - Total 3,200
    - Central bank — Liabilities:
      - Reserves (require 116) 140
      - Debt issued 500
      - Others 860
      - Other ST 1,250
      - CBDC 1,200 (+1,200)
      - Capital 500
      - Total 3,200

  - Case B (larger loan reduction shown second):
    - Consolidated banks — Assets:
      - Loans 5,520 (-480)
      - Cash 1,000
      - GB (gov. bond) 1,500
      - Securities 1,480 (+480)
      - Reserves 140 (Reserves (require 116) 140)
      - Others 250
      - Total 9,520
    - Consolidated banks — Liabilities:
      - Deposits 5,800 (-1,200)
      - Lending 1,470 (+720)
      - CB lending 1,470 (+720)
      - Debt issued 500
      - Others 860
      - Other ST 1,250
      - CBDC 1,200 (+1,200)
      - Capital 500
      - Total 9,520
    - Central bank — Assets:
      - CB lending 1,470 (+720)
      - Reserves 140
      - Total 3,200
    - Central bank — Liabilities:
      - Reserves (require 116) 140
      - Debt issued 500
      - Others 860
      - Other ST 1,250
      - CBDC 1,200 (+1,200)
      - Capital 500
      - Total 3,200

- Outcomes:
  - Banks reduce credit supply by more than in Scenario 4-A.
  - Banking sector becomes less profitable as funding costs increase across liabilities.

### Box 1: Changes in balance sheets following CBDC issuance through reserves and open market operations
- Core mechanics:
  - When households withdraw deposits for banknotes, reserves decline because banks use reserves to purchase banknotes.
  - When households purchase CBDC by reducing deposits, balance sheet changes are similar to increased cash holdings since CBDC is a central bank liability.
  - Initial effect: reduction in bank reserves on deposit with central bank; banks’ demand for reserves increases, putting upward pressure on interbank rates.
  - If central bank cannot adjust reserve requirement downward or does not operate a floor system supplying abundant reserves, it is likely to use open market operations to increase reserve supply.
  - Open market operations:
    - Temporary: central bank lending guaranteed by collateral (banks repay loan later).
    - Permanent: asset purchases (government bonds) that permanently increase reserves.

- Illustrative tables of balance sheet changes (preserved structure and labels):
  - Table A-1: CB assets: Reserves; CB liabilities: CBDC. Banks: Reserves (asset) and Deposits (liability). Households: Deposits and CBDC.
  - Table A-2: CB lending increases (CB asset) with corresponding Reserves (liability); Banks: Reserves (asset) and CB lending (liability).
  - Table A-3: CB Government bonds (GB) (asset) and Reserves (liability); Banks: Reserves (asset) and GB (liability).
  - Table A-4: CB CB lending (asset) and CBDC (liability); Banks Deposits and CB lending; HH Deposits and CBDC — corresponds to Scenario 2-A where banks have sufficient government bonds to collateralize borrowing.
  - Table A-5: CB CB lending and CBDC; Banks GB, Deposits, CB lending, Wholesale funding; NBFI GB and Wholesale funding; HH Deposits and CBDC — corresponds to commercial banks expanding balance sheets (Scenario 2-B).
  - Table A-6: CB GB (asset) and CBDC (liability); Banks Reserves, Deposits, Wholesale funding; NBFI GB and Wholesale lending; HH Deposits and CBDC — corresponds to asset purchase case (Scenario 3-A).
    - NOTE: NBFIs sell government bonds to central bank, neutralizing reserve decrease; banks use wholesale funding from NBFI to offset reserve decline.
  - Table A-7: CB Foreign GB ($) (asset) and CBDC (liability); Banks Deposits and Foreign wholesale funding ($) (liability); Foreign banks Foreign GB ($) and Foreign wholesale lending ($); HH Deposits and CBDC — corresponds to Scenario 3-B (foreign borrowing).
    - NOTE: Foreign bonds and wholesale funding assumed denominated in dollar ($). Central bank purchases dollars from commercial banks to buy dollar bonds; commercial banks sell dollars received as wholesale loan to central bank and receive local currency reserves needed to purchase CBDC.
  - Table A-8: CB GB and CBDC; Banks Loans and Deposits; HH Deposits and CBDC — corresponds to Scenario 4 where loans shrink and banks reduce wholesale funding.

- Additional mechanics notes:
  - In Scenario 3-B, central bank purchases local currency from commercial banks to buy foreign bonds, increasing total reserves (similar to unsterilized FX intervention).
  - In Scenario 4, steps include: households replace deposits with CBDC; banks reduce reserves/deposits to hand over CBDC; banks sell government bonds to central bank; banks increase wholesale funding to purchase government bonds from NBFI; when loan demand shrinks due to higher loan rates, banks lower wholesale funding, offsetting some initial wholesale borrowing.

### Annex II: Quantitative indicators — cross-country variation and implications
- Variation and relevance:
  - Countries differ significantly in initial conditions that determine which scenario may arise (e.g., low financial inclusion may lead CBDC to replace cash rather than deposits).
  - Where banking sectors hold ample reserves, issuance of CBDC funded by drawdown of existing reserves may have limited adverse financial stability implications.

- Indicators and interpretations (figures referenced; 5-year averages 2017–21):
  - Cash in circulation (CIC)-to-GDP ratio versus share of transfer deposits in total banking sector liabilities (Annex Figure 1):
    - Indicates likelihood of CBDC replacing cash (upper left region: high cash-to-deposit ratio) versus replacing deposits (bottom right region).
    - Other elements (household payment preferences, CBDC design) also affect outcome.
  - Share of reserves in banks’ balance sheets (Annex Figure 2):
    - Banking sectors on left side of chart more likely to have sufficient excess reserves to reduce when deposits are replaced by CBDC.
    - Sectors on right side may not hold sufficient reserves and would need other funding sources (central bank lending, wholesale funding) or reduce assets.
    - Note: reserve share used; does not necessarily highlight excess reserves because reserve requirements differ across countries (cross-country reserve requirement data not readily available).
  - Interest margins across banking sectors (Annex Figure 3):
    - When interest margins are tight (high competition, less profitability), pass-through to loan rates from a rise in funding cost may be larger because banks lack margin to absorb expense increases.
    - In such cases, provision of credit is likely to shrink more sharply as loan supply shifts in.
    - Interest margin can indicate likelihood of risk-shifting incentives as profits are squeezed; banks may invest in riskier projects or exit in highly competitive environments.

_International Monetary Fund — IMF WORKING PAPERS: Central Bank Digital Currencies and Financial Stability: Balance Sheet Analysis and Policy Choices_

### References

### References

### Works on CBDC design, adoption, and implications
- Adalid, R., Á. Álvarez-Blázquez, K. Assenmacher, L. Burlon, M. Dimou, C. López-Quiles, N.M. Fuentes, B. Meller, M.A. Muñoz, P. Radulova, C. Rodriguez d’Acri, T. Shakir G. Šílová, O. Soons, and A.V. Veghazy, 2022. “Central Bank Digital Currency and Bank Intermediation.” ECB Occasional Paper No. 293.
- Agur, I., A. Ari, and G. Dell’Ariccia. 2022. “Designing Central Bank Digital Currencies.” Journal of Monetary Economics, 125, 62–79.
- Andolfatto, D. 2021. “Assessing the Impact of Central Bank Digital Currency on Private Banks.” The Economic Journal, 131(634), 525–40.
- Armas, A., and Singh, M. M. 2022. “Digital Money and Central Banks Balance Sheet.” IMF Working Paper No. 236.
- Auer, R., Banka, H, Boakye-Adjei, N., Faragallah, A., Frost, J., Natarajan, H., and Prenio, J. 2022. “Central Bank Digital Currencies: A New Tool in the Financial Inclusion Toolkit?” FSI Insights No 41, April.
- Auer, S., Branzoli, N., Ferrero, G., Ilari, A., Palazzo, F., & Rainone, E. 2024. “CBDC and the Banking System.” Bank of Italy, Economic Research and International Relations Area (No. 829).
- Bindseil, U. 2020. “Tiered CBDC and the Financial System.” ECB Working Paper Series No. 2351. January.
- BIS. 2021a. “Central Bank Digital Currencies: Foundational Principles and Core Features.” Report no 1 on CBDC publication series with collaborations from a group of central banks.
- BIS. 2021b. “Central Bank Digital Currencies: Financial Stability Implications.” Report no. 4 on CBDC publication series with collaborations from a group of central banks.
- BIS 2021c. “Central Bank Digital Currencies: User Needs and Adoption.” Report no. 3 on CBDC publication series with collaborations from a group of central banks.
- Chiu, J., S. M. Davoodalhosseini, J. Jiang, and Y. Zhu. 2023. “Bank Market Power and Central Bank Digital Currency: Theory and Quantitative Assessment.” Journal of Political Economy, Vol. 131, No. 5, 1213–48.
- Choi, K. and K. Rhee. 2022. “What Doesn’t Kill You Only Makes You Riskier: Impacts of CBDC on Banking Stability.” October. Unpublished.
- Juks, R. 2018. “When a Central Bank Digital Currency Meets Private Money: The Effects of an E-Krona on Banks.” Sveriges Riksbank Economic Review, 3, 79–99.
- Keister, T. and D. Sanches. 2023. “Should Central Banks Issue Digital Currency?” The Review of Economic Studies, 90(1), 404–31.
- Keister, T. and C. Monnet. 2022. “Central Bank Digital Currency: Stability and Information.” Journal of Economic Dynamics and Control, 142.
- Li, J. 2022. “Predicting the Demand for Central Bank Digital Currency: A Structural Analysis with Survey Data.” Journal of Monetary Economics. Volume 134, March, pp. 73-85.
- Lannquist, A. and B. Tan. 2023. “Central Bank Digital Currency's Role in Promoting Financial Inclusion.” FinTech Notes 11.
- Mancini-Griffoli, T., M.S.M. Peria, I. Agur, A. Ari, J. Kiff, A. Popescu, and C. Rochon. 2018. “Casting Light on Central Bank Digital Currency.” IMF Staff Discussion Note No. 8.
- Meller, B. and Soons, O., 2023. “Know your (holding) limits: CBDC, financial stability and central bank reliance,” ECB Occasional Paper Series No 326.
- Soderberg, G., M. Bechara, W. Bossu, W., N. Che, S. Davidovich, J. Kiff, I. Lukonga, T. Mancini-Griffoli, T. Sun, and A. Yoshinaga. 2022. “Behind the Scenes of Central Bank Digital Currency: Emerging Trends, Insights, and Policy Lessons”. IMF FinTech Note No. 4.
- Stanley, A. 2022. “The Ascent of CBDCs.” IMF Finance & Development, September. p. 48-49.
- Zamora-Pérez, A., E. Coschignano, and L. Barreiro. 2022. “Ensuring Adoption of Central Bank Digital Currencies – An Easy Task or a Gordian Knot?” ECB Occasional Paper 307.
- Gross, M. and E. Letizia. 2023. “ To Demand or Not to Demand: On Quantifying the Future Appetite for CBDC.” IMF Policy Paper No. 9.
- Bijlsma, M., C. van der Cruijsen, N. Jonker, and J. Reijerink. 2024. “What Triggers Consumer Adoption of CBDC?” Journal of Financial Service Research 65, pp. 1-40.
- Burlon, L., C. Montes-Galdon, M. Muñoz, and F. Smets. 2022. “The Optimal Quantity of CBDC in a Bank-Based Economy.” ECB Working Paper No. 2689. July.
- Bucholtz, Z., N., Michelson, B. Ettinger, and Y. Soffer. 2022. “Digital Shekel: Potential Effects of Issuing a Digital Shekel on the Israeli Banking System.” The Bank of Israel Steering Committee on the Potential Issuance of a Digital Shekel. March.
- Carapella, F., J. Chang, S. Infante, M. Leistra, A. Lubis, and A.P. Vardoulakis. 2024. “Financial Stability Implications of CBDC.” Finance and Economics Discussion Series No. 21. Washington: Board of Governors of the Federal Reserve System.
- Chang, H., F. Grinberg, L. Gornicka, M. Miccoli, and B. Tan. 2023. “Central Bank Digital Currency and Bank Disintermediation in a Portfolio Choice Model.” IMF Working Paper No. 2023/236.
- Infante, S., K. Kim, A. F. Silva, and R. J. Tetlow. 2022. “The Macroeconomic Implications of CBDC: A Review of the Literature.” Finance and Economics Discussion Series No. 2022-76. Federal Reserve Board.
- Infante, S., K. Kim, A. F. Silva, and R. J. Tetlow. 2023. ”Retail Central Bank Digital Currencies: Implications for Banking and Financial Stability.” Finance and Economics Discussion Series No. 2023-72. Federal Reserve Board.
- He, M. D., A. Kokenyne, M. Griffoli, M. Miccoli, M. Olafsson, G. Soderberg, and H. Tourpe. 2023. “Capital Flow Management Measures in the Digital Age (2): Design Choices for Central Bank Digital Currency.” IMF Fintech Notes No. 9.
- Williamson, S. D. (2022): “Central bank digital currency and flight to safety.” Journal of Economic Dynamics & Control 142
- Whited, T. M., Y. Wu, and K. Xiao. 2023. “Will Central Bank Digital Currency Disintermediate Banks?” Working paper.

### Banking, financial intermediation, stability, and crisis literature
- Allen, F., and Gale, D. 2004. “Competition and Financial Stability.” Journal of Money, Credit and Banking, Vol. 36, No. 3. 453–80.
- Berlin, M. and L. Mester. 1999. “Deposits and Relationship Lending.” Review of Financial Studies 12, 579–607.
- Bernanke, B.S. 1983. “Irreversibility, Uncertainty, and Cyclical Investment.” The Quarterly Journal of Economics, 98(1), pp.85–106.
- Berger, references not present beyond supplied content — [note: skip]
- Bitter, L. 2023. “Banking Crises under a Central Bank Digital Currency (CBDC).” Discussion Paper, No. 426, Ludwig-Maximilians-Universität München und Humboldt Universität zu Berlin.
- Board, F. S. 2009. “Report to G20 Finance Ministers and Governors Guidance to Assess the Systemic Importance of Financial Institutions, Markets and Instruments: Initial Considerations.” Wayamba Journal of Management, 10(1).
- Brooks, S. 2021. “Revisiting the Monetary Sovereignty Rationale for CBDCs.” Bank of Canada Staff Discussion Paper No. 2021–17.
- Brunnermeier, M. K. and D. Niepelt. 2019. “On the Equivalence of Private and Public Money.” Journal of Monetary Economics, 106, 27–41.
- Choi, D. B., J. A. Santos, and T. Yorulmazer. 2021. “A Theory of Collateral for the Lender of Last Resort.” Review of Finance, 25 (4), 973–96.
- Cunliffe, J. 2023. “The Digital Pound.” Speech given at UK Finance, published on February 7.
- Dell'Ariccia, G., L. Laeven, and R. Marquez. 2014. “Real Interest Rates, Leverage, and Bank Risk-Taking.” Journal of Economic Theory, 149, 65–99.
- Dell’Ariccia, G., D. Igan, L. Laeven, and H. Tong, 2016. “Credit Booms and Macrofinancial Stability.” Economic Policy, 31(86), pp.299–55.
- Dell'Ariccia, G., C. Ferreira, N. Jenkinson, M.L. Laeven, A. Martin, M.C. Minoiu, and A. Popov. 2017. “Managing the Sovereign-Bank Nexus”, IMF Departmental Paper No. 16.
- Diamond, D.W. and P.H. Dybvig. 1983. “Bank Runs, Deposit Insurance, and Liquidity.” Journal of Political Economy, 91(3), pp.401-419.
- Donaldson, J. R., G. Piacentino, A. Thakor. 2018. “Warehouse Banking.” Journal of Financial Economics 129, 250–67.
- Gerali, A. and F. Passacantando. 2007. “The Loss of Confidence of Bank Money in the Great Depression.” Paper presented at Payments and Monetary and Financial Stability, ECB-Bank of England Conference 12-13 November 2007, in The Economics of Large-Value Payments and Settlements. Theory and Policy Issues for Central Banks, M. Manning, E. Nier, and J. Schanz (ed.). Oxford University Press. New York. November 2009.
- Grodecka-Messi, A. and X. Zhang. 2023. “Private Bank Money vs Central Bank Money: A Historical Lesson for CBDC Introduction.” Journal of Economic Dynamics and Control 154.
- Kahn, C.M. 2022, “Should the United States Issue a Central Bank Digital Currency? Lessons from Abroad.” Federal Reserve Bank of Atlanta Policy Hub, June.
- Khwaja, A. I., and A. Mian. 2008. “Tracing the Impact of Bank Liquidity Shocks: Evidence from an Emerging Market.” The American Economic Review, 98(4), 1413-42, September.
- Keeley, M. C. 1990. “Deposit Insurance, Risk, and Market Power in Banking”, The American Economic Review, 80 (5), 1183–200. December.
- King, M. 2016. The End of Alchemy: Money, Banking, and the Future of the Global Economy. WW Norton & Company.
- Mester, L. J., L. I. Nakamura, and M. Renault. 2007. “Transactions Accounts and Loan Monitoring.” The Review of Financial Studies, 20, 529–56.
- Merrouche, O. and E.W. Nier, 2012. “Payment Systems, Inside Money and Financial Intermediation.” Journal of Financial Intermediation, 21 (3), 359–82.
- Peek, Joe, and E.S. Rosengren. 2000. “Collateral Damage: Effects of the Japanese Bank Crisis on Real Activity in the United States.” The American Economic Review, 90 (1), 30–45.
- Puri, M., J. Rocholl, and S. Steffen. 2017. "What Do a Million Observations Have to Say about Loan Defaults? Opening The Black Box of Relationships." Journal of Financial Intermediation, 31, 1–15.
- Roberts, J. M. 2018. “An Estimate of the Long-Term Neutral Rate of Interest,” Fed Notes September 05.
- Sanches, D. and T. Keister. 2021. “Should Central Banks Issue Digital Currency?” Federal Reserve Bank of Philadelphia Working Paper No. 37.
- Seidemann, W. 2021. “The Role of Commercial Central Banks in CBDCs”. The Global Treasurer, November 15.
- Stiroh, K.J. 2004. "Diversification in Banking: Is Noninterest Income the Answer? " Journal of Money, Credit and Banking, Vol. 36, No. 5, October, 853–82.
- Surti, J. 2021. “Risk, Return and the Search for Yield.” IMF Finance & Development, June. p. 48-49.
- Veyrune, M. R. M., S. Guo, and J. Alwazir. 2019. “Autonomous Factor Forecast Quality: The Case of the Eurosystem.” IMF Working Papers No. 296.
- Whited, T. M., Y. Wu, and K. Xiao. 2023. “Will Central Bank Digital Currency Disintermediate Banks?” Working paper.

### Payment systems, stablecoins, and fintech impacts
- Aldasaro, I., W. Huang, and E. Kemp. 2020. “Cross-border Links between Banks and Non-bank Financial Institutions.” BIS Quarterly Review, September.
- Bagnall, J., D. Bounie, K.P. Huynh, A. Kosse, T. Schmidt, S. Schuh, and H. Stix. 2014. “Consumer Cash Usage. A Cross-Country Comparison with Payment Diary Survey Data.” ECB Working Paper 1685, June.
- BIS. 2016. “Fast Payments: Enhancing the Speed and Availability of Retail Payments.” Basel, Switzerland: Committee on Payments and Market Infrastructures.
- Garratt, R., H.S. Shin. 2023. “Stablecoins versus Tokenised Deposits: Implications for the Singleness of Money.” BIS Bulletin No. 73.
- Ghosh, P., B. Vallee, and Y. Zeng. 2021. “Fintech Lending and Cashless Payments.” In Proceedings of Paris December 2021 Finance Meeting EUROFIDAI-ESSEC.
- Gowrisankaran, G. and J. Stavins. 2004. “Network Externalities and Technology Adoption: Lessons from Electronic Payments.” The RAND Journal of Economics, Vol. 35, No. 2, pp. 260-276.
- Hau, H., Y. Huang, H. Shan, and Z. Sheng. 2019. “How FinTech Enters China's Credit Market.” American Economic Review Papers and Proceedings, 109, 60–64.
- Liao, G. Y. and J. Caramichael. 2022. “Stablecoins: Growth Potential and Impact on Banking.” International Finance Discussion Papers No. 1334. Board of Governors of the Federal Reserve System.
- MacDonald, C., and L. Zhao, 2022. “ Stablecoins and their Risks to Financial Stability.” Bank of Canada Staff Discussion Paper No. 20.
- McKinsey. 2019. Global Payments Report 2019: Amid Sustained Growth, Accelerating Challenges Demand Bold Actions. Technical Report. September.
- Parlour, C. A., U. Rajan, and H. Zhu. 2022. “When FinTech Competes for Payment Flows.” The Review of Financial Studies, 35(11), 4985–5024.
- Stanley, A. 2022. “The Ascent of CBDCs.” IMF Finance & Development, September. p. 48-49.

### International organizations, standards, and policy reports
- Banco Central do Brazil. 2022. Financial Stability Report, Vol. 21, no. 2. https://www.bcb.gov.br/content/publications/financialstabilityreport/202211/fsrFullRep.pdf
- Bank of England. 2023. “Regulatory Regime for Systemic Payment Systems using Stablecoins and Related Service Providers.” Bank of England Discussion Paper.
- BIS. 2013. “Basel III: The Liquidity Coverage Ratio and Liquidity Risk Monitoring Tools.” Basel Committee on Banking Supervision. Bank for International Settlements. January. Basel.
- BIS. 2014a. “Basel III: Leverage Ratio Framework and Disclosure Requirements.” Bank for International Settlements. January. Basel.
- BIS. 2014b. “Basel III: The Net Stable Funding Ratio. Basel Committee on Banking Supervision.” Bank for International Settlements. October. Basel.
- BIS 2023a. “  Blueprint for the Future Monetary System: Improving the Old, Enabling the New.” BIS Annual Economic Report Chapter 3. Basel. June.
- BIS 2023b. “Central Bank Digital Currency (CBDC) Information Security and Operational Risks to Central Banks.” Report by the Consultative Group on Risk Management (CGRM) established at the BIS Representative Office for the Americas. November.
- European Central Bank. 2019. “Exploring Anonymity in Central Bank Digital Currencies.” IN FOCUS issue no. 4.
- European Central Bank. 2022. Progress on the Investigation Phase of a Digital Euro. Second progress report, December 21.
- European Commission. 2023. “European Digital Identity.” Website of the European Commission. https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/europe-fit -digital-age/european-digital-identity_en.
- Financial Stability Board. 2020. Holistic Review of the March Market Turmoil. FSB. November.
- Financial Stability Board. 2021. Policy Proposals to Enhance Money Market Fund Resilience. FSB.
- Financial Stability Board. 2022. Assessment of the Effectiveness of the FSB’s 2017 Recommendations on Liquidity Mismatch in Open-Ended Funds. FSB.
- G7 Finance Ministers and Central Bank Governors’ Statement on Central Bank Digital Currencies (CBDCs) and Digital Payments – 13 October 2021.
- G30 Working Group on the 2023 Banking Crisis. 2024. “Bank Failures and Contagion: Lender of Last Resort, Liquidity, and Risk Management.” Special Report.
- International Organization of Securities Commissions. 2012. “Policy Recommendations for Money Market Funds”. FR07/12 Policy Recommendations for Money Market Funds (iosco.org)
- International Organization of Securities Commissions. 2018. “Recommendations for Liquidity Risk Management for Collective Investment Schemes” FR01/2018 Recommendations for Liquidity Risk Management for Collective Investment Schemes (iosco.org)
- Sveriges Riksbank. 2022. “E-krona Pilot Phase 2.” E-krona Report, April.
- U.S. Government Accountability Office. 2023. "Bank Regulation: Preliminary Review of Agency Actions Related to March 2023 Bank Failures." GAO-23-106736. April.
- Federal Deposit Insurance Corporation. 2023a. "FDIC Supervision of Signature Bank.” April.
- Federal Deposit Insurance Corporation. 2023b. “Options for Deposit Insurance Reform.” April.
- Federal Reserve Board. 2023. “Review of the Federal Reserve’s Supervision and Regulation of Silicon Valley Bank.” April.

### Macroprudential, funds, and market functioning
- BIS. 2016. “Fast Payments: Enhancing the Speed and Availability of Retail Payments.” Basel, Switzerland: Committee on Payments and Market Infrastructures.
- Financial Stability Board. 2020. Holistic Review of the March Market Turmoil. FSB. November.
- Forbes, K., C. Friedrich, D. Reinhardt. 2022. “Stress Relief? Financial Structures and Resilience to the COVID Shock.” Unpublished.
- International Monetary Fund (IMF). 2013. “Key Aspects of Macroprudential Policy – Background Paper: Key Aspects of Macroprudential Policy--Background Paper; IMF Policy Paper; June 10, 2013.
- IMF. 2021. “Investment Funds and Financial Stability, Policy Considerations.” Monetary and Capital Markets Departmental Paper 18.
- IMF. 2022a. Global Financial Stability Review, Chapter 2, “The Sovereign-Bank Nexus in Emerging Markets: A Risky Embrace.” International Monetary Fund.
- IMF. 2022b. Global Financial Stability Report, Chapter 3 “Asset Price Fragility in Times of Stress: The Role of Open-End Investment Funds.” International Monetary Fund.
- IMF. 2022c. "Regulating the Crypto Ecosystem: The Case of Stablecoins and Arrangements”, FinTech Notes No. 8.
- IMF. 2022d. “Nigeria: 2022 Article IV Consultation Staff Report.” Country Report No. 93.
- IMF. 2023a. Global Financial Stability Report, “Nonbank Financial Intermediaries: Vulnerabilities amid Tighter Financial Conditions”, April.
- IMF. 2023b. “Elements of Effective Policies for Crypto Assets” IMF Policy Paper No. 2023/004, February 23.
- IMF. 2023c. “IMF Approach to Central Bank Digital Currency Capacity Development” IMF Policy Paper No. 2023/016, April 11.
- Infante, S., K. Kim, A. F. Silva, and R. J. Tetlow. 2023. ”Retail Central Bank Digital Currencies: Implications for Banking and Financial Stability.” Finance and Economics Discussion Series No. 2023-72. Federal Reserve Board.
- Liao, G. Y. and J. Caramichael. 2022. “Stablecoins: Growth Potential and Impact on Banking.” International Finance Discussion Papers No. 1334. Board of Governors of the Federal Reserve System.
- MacDonald, C., and L. Zhao, 2022. “ Stablecoins and their Risks to Financial Stability.” Bank of Canada Staff Discussion Paper No. 20.
- Soderberg, G., et al. 2022. “Behind the Scenes of Central Bank Digital Currency: Emerging Trends, Insights, and Policy Lessons”. IMF FinTech Note No. 4.

*Central Bank Digital Currencies and Financial Stability: Balance Sheet Analysis and Policy Choices Working Paper No. WP/2024/226*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024226-print-pdf.pdf_