## Annex A

## Source details

**Canonical URL:** [Annex A](https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022114-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2022/english/wpiea2022114-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2022/english/wpiea2022114-print-pdf.pdf.json)

---

### Data sources and scope
- By announcement: National official sources (Policy announcements, budget amendments, official reports, central bank balance sheets, etc.) and Online sources (News reports, private-sector reports, IMF Article IV reports, reports by regional organizations, etc.).
- Coverage: Extraordinary policy actions announced during 2020 and directly or indirectly related to Covid-19 (criterion of relevance).
- Unit of observation: Announcement-level. A measure must be announced to be recorded.
- Implementing authority: Measures implemented at the central government level; measures by local governments or foreign entities are generally not included.
- European-level measures: Allocated to individual member states; duplicated for all EA member states and EU member states in the country-level tracker. Next Generation EU recorded as a fiscal announcement but without sizes.

### Criteria for recording policy measures
- Timing:
  - Use the earliest official announcement that offers details on the nature of the measure.
  - Tentative proposals with significant uncertainties are not recorded.
  - If operational details (e.g., size) are announced later, the first announcement determines timing and later announcements are used as supplementary information (example: March 20 initial announcement and size announced one week later—use March 20 as announcement date and impute size from later announcement).
- Pre-existing measures: Not counted unless later announcements update or expand them in response to Covid-19; record the relevant changes rather than pre-existing policies.
- Automatic stabilizers: Not recorded unless there is a specific government announcement.
- Unannounced actions: Not recorded (e.g., FX interventions without announcement or news report are excluded).
- No actions: Announcements of no action do not constitute a policy measure.
- Extensions and rollbacks: Coded as new measures; dataset constructs dummies flagged as “+1” for extensions and “-1” for rollbacks. In coding sizes of extensions, only incremental sizes are incorporated to avoid double counting. Automatic expirations are not counted.

### Classification, coding, and size conventions
- Four broad policy categories: fiscal, monetary, prudential, and other.
- Direction dummies:
  - For each policy type a dummy takes value +1, 0, or -1.
  - “0” means announcement does not correspond to the policy type.
  - “+1” means loosening policy.
  - “-1” means tightening policy.
  - Exceptions and special rules:
    - Dividend restrictions assigned “+1” (considered loosening because intended to positively impact credit by preserving capital).
    - FX interventions: “+1” intended to strengthen or stabilize exchange rates; “-1” intended to weaken exchange rates.
    - Discretionary rollbacks given opposite sign of original measures.
    - NBFI and market-based measures under “Other” category only take +1 or 0.
- Size recording:
  - Announced sizes: All sizes are announced sizes or early estimates. When not available, use official or unofficial estimates published after announcements, minimizing reliance on information not available at announcement time.
  - Missing values: Size recorded as missing when no information available or size cannot be quantified.
  - Budgetary impact vs program sizes:
    - Above-the-line fiscal measures (grants, tax reliefs, tax deferrals): record budgetary impact.
    - Below-the-line (equity participations and public loans) and contingent (public guarantees): record program sizes.
    - When a loan (guarantee) program also involves a capital injection, track both the loan (guarantee) program and the equity injection under their respective policy types.
  - Actual sizes: When announced sizes unavailable, collect actual sizes for asset purchase programs and credit facilities (subset of lending operations). The actual size of a measure is calculated as the maximum increase between its announced starting date and the earlier of (i) its next extension, or (ii) the end of Q1 2021.
  - Horizon: Near-term size focus. For fiscal measures, near term defined as expenditure in 2020 and 2021 (estimate portion allocated to 2020 and 2021 if long-term policy announced). For asset purchases and credit facilities, near term includes up to Q1 2021.
- Multi-policy fiscal announcements: When combined size known but cannot be allocated to individual policies, record policies in same line and report combined size; sizes of individual policies recorded as missing.

### Policy types and size availability (high-level)
- Fiscal policy types with size information (Y): Grants, Tax reliefs, Tax deferrals, Public loans, Public guarantees, Equity participations.
- Monetary policy types with size information (Y): Lending operations*, Asset purchases*, Central bank swap lines, Foreign exchange Interventions.
  - Note: Y* denotes that both announced and actual sizes are recorded. For lending operations, only the credit facility subset has announced and actual sizes.
- Prudential policy types: Mixed size availability; Macroprudential buffers Y, many other prudential types size N.
- Other category: Many measures (Non-bank financial institutions, Market-based measures, Moratoria, Reporting requirements) generally have Size: N.

### Overview of the tracker (coverage and timing)
- Database contains 5,462 policies announced by governments throughout 2020.
- Country coverage: 74 countries.
- Timing concentration:
  - For both AEs and EMDEs, policies highly concentrated in March and April 2020.
  - Over March–April 2020, the median number of policies announced was 76 for AEs and 28 for EMDEs.
  - From May onwards, the median dropped to 29 for AEs and 17 for EMDEs.
  - March 12 (18) was the single day with the most fiscal, monetary, and prudential policies announced among AEs (EMDEs).
- Stylized facts:
  - Countries tended to introduce different types of policies simultaneously, leading to positive and statistically significant correlations across policy types.
  - Standalone policy announcements are rare, complicating isolation of specific policy impacts.

### Fiscal policy findings
- Frequency and types:
  - Grants were the most common fiscal policy measure in both AEs and EMDEs in 2020.
  - In AEs, public loans and guarantees were the next most frequently used instruments after grants.
  - In EMDEs, tax reliefs were used mostly after grants.
- Magnitude and cross-country distribution:
  - Size of fiscal measures was larger among AEs relative to EMDEs.
  - Country-level medians:
    - Median size of all fiscal measures combined in AEs: 15.2 percent of GDP.
    - Median size of all fiscal measures combined in EMDEs: 4.1 percent of GDP.
  - Dispersion:
    - Standard deviation for AEs: 7.5 percentage points.
    - Standard deviation for EMs: 5.6 percentage points.
  - Example policy: Australia government announced a $17.6 billion economic stimulus package on March 12.

### Monetary policy findings
- Instruments and usage:
  - Advanced economies (AEs) relied more on asset purchases (given policy rates at or close to effective lower bound).
  - EMDEs introduced more changes in interest rates and reserve requirements.
  - Credit and liquidity injections were widely used by both groups.
- Asset purchases:
  - Median cumulative asset purchases among AEs: 17.7 percent of GDP.
  - Standard deviation for AEs asset purchases: 6.5 percentage points.
  - Median cumulative asset purchases among EMDEs: 1.3 percent of GDP.
  - Standard deviation for EMDEs asset purchases: 2.7 percentage points.
- Policy rate cuts:
  - Median (mean) policy rate cut by AEs in 2020: 1.5 (1.1) percentage points.
    - These AEs are: Australia, Canada, Czech Republic, Hong Kong, Israel, South Korea, Norway, the United Kingdom, and the United States.
  - Median (mean) policy rate cut by EMDEs in 2020: 1.25 (1.9) percentage points.
  - Positive correlations across monetary measures indicate they were applied in tandem during the pandemic.

### Prudential policy findings
- Aggregate counts in 2020:
  - AEs announced 400 capital-related prudential measures and 348 non-capital prudential measures.
  - EMDEs announced 181 capital-related prudential measures and 202 non-capital prudential measures.
- Capital-related measures:
  - Changes in buffers and dividend restrictions were introduced, but the relaxation of other capital-related measures was more prevalent, especially among AEs.
  - Relaxation of provisioning requirements was frequently adopted among EMDEs.
- Non-capital measures:
  - AEs frequently relaxed supervisory expectations and reporting requirements.
- Correlations:
  - Different types of prudential policies were introduced simultaneously, leading to positive and significant correlations across policies.

### Key methodological appendices (high-level)
- Implementing authorities:
  - Above-the-line measures generally considered only when announced by central governments; local governments not directly covered except via extraordinary transfers from central governments.
  - Equity participations, public loans, and public guarantees can be provided by governments or state-owned entities; support from central banks or international organizations is excluded except specific cases (European Guarantee Fund).
  - Monetary measures should be implemented by central banks; government purchases of securities on secondary markets are classified as fiscal (equity or public loans) not as asset purchase programs.
- Special classification issues:
  - Distinguish within central bank lending operations between credit facilities and market liquidity measures based on stated aims and maturity (lending operations maturing in one month generally classified as market liquidity measures).
  - When central bank lending intertwines with public loans, classification depends on implementing authority and on-lending specifics.
  - Equity injections associated with credit-related measures are recorded separately under the “equity” variable.
  - European-level measures allocation rules include specific treatment for ECB asset purchases (Eurosystem capital key as of February 1 2020), duplication of EA/EU prudential measures, and recording Next Generation EU as fiscal announcements without sizes.
- Country coverage and lists:
  - AEs: 30 countries listed (1 Australia … 30 United States).
  - EMDEs: 44 countries listed (31 Argentina … 74 Vietnam).
- Policy definitions and size availability:
  - Appendix Table A.2 enumerates 28 policy types (1–6 Fiscal; 7–14 Monetary; 15–25 Prudential; 26–28 Other) with precise definitions and Size: Y/N/Y* annotations.
- Database variables:
  - Appendix Table A.3 lists all database variables including identifiers (ID, country, Ifs, income_group, region, NGDP_2019, year, quarter, month, date, description, link, var_summary), policy-specific size variables (grant_size, grant_gdp, APP_size, APP_gdp, CF_size, CF_gdp, pol_rate_size, RR_LCU_size, RR_FX_size, swap_line_size, etc.), prudential variables (pru_CCyB, pru_CCoB, pru_SyRB, pru_reporting_requirements, etc.), other policy variables (moratorium_coverage, moratorium_duration_min, moratorium_duration_max, moratorium_bank), and supplementary flags (extension, rollback, european, BIS_ID, ESRB_ID, YPFS_ID).
  - Conversion note: When converting currency units and calculating sizes as percent of GDP, use nominal GDP and end-of-period exchange rates in 2019 from the October 2021 World Economic Outlook Database.

### Conclusions (key takeaways)
- The announcement-level policy tracker for 74 countries contains 5,462 policies for 2020 and is more comprehensive and accurate for the countries in the dataset than existing trackers for that period.
- Most policy announcements occurred early in the pandemic (concentrated in March and April 2020).
- Announcements were highly correlated across policy types, making isolation of specific policy impacts challenging.
- Fiscal measures were significantly larger in AEs than in EMDEs, though dispersion within AEs and EMDEs is large.
- Central banks in EMDEs primarily cut interest rates and reserve requirements; central banks in AEs relied on asset purchase programs for monetary stimulus.
- Asset purchases were strikingly larger among AEs wherever used.
- Both AEs and EMDEs announced capital and non-capital prudential measures; buffer relaxations occurred in some AEs, other capital-related relaxations were widely implemented in both groups, provisioning requirement relaxations were more common among EMDEs, and supervisory expectations and reporting requirements were relaxed in both groups.

*Source: Annex A, wpiea2022114-print-pdf (IMF Working Paper).*

### Annex A

### Annex A

### Data sources and scope
- By announcement: National official sources (Policy announcements, budget amendments, official reports, central bank balance sheets, etc.) and Online sources (News reports, private-sector reports, IMF Article IV reports, reports by regional organizations, etc.).
- Coverage: Extraordinary policy actions announced during 2020 and directly or indirectly related to Covid-19 (criterion of relevance).
- Unit of observation: Announcement-level. A measure must be announced to be recorded.
- Implementing authority: Measures implemented at the central government level; measures by local governments or foreign entities are generally not included.
- European-level measures: Allocated to individual member states; duplicated for all EA member states and EU member states in the country-level tracker. Next Generation EU recorded as a fiscal announcement but without sizes.

### Criteria for recording policy measures
- Timing:
  - Use the earliest official announcement that offers details on the nature of the measure.
  - Tentative proposals with significant uncertainties are not recorded.
  - If operational details (e.g., size) are announced later, the first announcement determines timing and later announcements are used as supplementary information (example: March 20 initial announcement and size announced one week later—use March 20 as announcement date and impute size from later announcement).
- Pre-existing measures: Not counted unless later announcements update or expand them in response to Covid-19; record the relevant changes rather than pre-existing policies.
- Automatic stabilizers: Not recorded unless there is a specific government announcement.
- Unannounced actions: Not recorded (e.g., FX interventions without announcement or news report are excluded).
- No actions: Announcements of no action do not constitute a policy measure.
- Extensions and rollbacks: Coded as new measures; dataset constructs dummies flagged as “+1” for extensions and “-1” for rollbacks. In coding sizes of extensions, only incremental sizes are incorporated to avoid double counting. Automatic expirations are not counted.

### Classification, coding, and size conventions
- Four broad policy categories: fiscal, monetary, prudential, and other. (Table 2 lists individual policy types; Appendix Table A.2 provides exact definitions.)
- Direction dummies: For each policy type a dummy takes value +1, 0, or -1.
  - “0” means announcement does not correspond to the policy type.
  - “+1” means loosening policy.
  - “-1” means tightening policy.
  - Exceptions:
    - Dividend restrictions assigned “+1” (considered loosening because intended to positively impact credit by preserving capital).
    - FX interventions: “+1” intended to strengthen or stabilize exchange rates; “-1” intended to weaken exchange rates.
    - Discretionary rollbacks given opposite sign of original measures.
    - NBFI and market-based measures under “Other” category only take +1 or 0 (no loosening/tightening distinction).
- Size recording:
  - Announced sizes: All sizes are announced sizes or early estimates. When not available, use official or unofficial estimates published after announcements, minimizing reliance on information not available at announcement time.
  - Missing values: Size recorded as missing when no information available or size cannot be quantified (example: delaying a scheduled increase in capital conservation buffer without changing current requirement).
  - Budgetary impact vs program sizes:
    - Above-the-line fiscal measures (grants, tax reliefs, tax deferrals): record budgetary impact.
    - Below-the-line (equity participations and public loans) and contingent (public guarantees): record program sizes (amounts of capital injections for equity measures and loan amounts for public loans and guarantees).
    - When a loan (guarantee) program also involves a capital injection, track both the loan (guarantee) program and the equity injection under their respective policy types.
  - Actual sizes: When announced sizes unavailable, collect actual sizes for asset purchase programs and credit facilities (subset of lending operations). The actual size of a measure is calculated as the maximum increase between its announced starting date and the earlier of (i) its next extension, or (ii) the end of Q1 2021. Example: if an asset purchase program announced in March 2020 is extended in September 2020 and February 2021, actual size of March 2020 announcement = maximum increase from March to September 2020; September 2020 announcement actual size = maximum increase from September 2020 to February 2021. February 2021 announcement not recorded if it falls out of sample period.
  - Horizon: Near-term size focus. For fiscal measures, near term defined as expenditure in 2020 and 2021 (estimate portion allocated to 2020 and 2021 if long-term policy announced). For asset purchases and credit facilities, near term includes up to Q1 2021.
- Multi-policy fiscal announcements: When combined size known but cannot be allocated to individual policies, record policies in same line and report combined size; sizes of individual policies recorded as missing.

### Policy types and size availability (Table 2 highlights)
- Fiscal policy types with size information (Y): Grants, Tax reliefs, Tax deferrals, Public loans, Public guarantees, Equity participations.
- Monetary policy types with size information (Y): Lending operations*, Asset purchases*, Central bank swap lines, Foreign exchange Interventions.
  - Note: Y* denotes that both announced and actual sizes are recorded. For lending operations, only the credit facility subset has announced and actual sizes.
- Prudential policy types: Capital-related requirements (size N for most capital requirements), Reserve requirements (local currency) N, Reserve requirements (foreign currency) N, Macroprudential buffers Y, Liquidity requirements N, Buffer usability N, Capital-related measures may have mixed size availability as noted in table.
- Other category: Non-bank financial institutions N, Market-based measures N, Moratoria N, Reporting requirements N, etc.
- Notes: Y (N) indicates that size information is (not) available for the corresponding policy.

### Overview of the tracker (coverage and timing)
- Database contains 5,462 policies announced by governments throughout 2020.
- Country coverage: 74 countries.
- Timing concentration:
  - For both AEs and EMDEs, policies highly concentrated in March and April 2020.
  - Over March–April 2020, the median number of policies announced was 76 for AEs and 28 for EMDEs.
  - From May onwards, the median dropped to 29 for AEs and 17 for EMDEs.
  - March 12 (18) was the single day with the most fiscal, monetary, and prudential policies announced among AEs (EMDEs).
- Stylized facts:
  - Countries tended to introduce different types of policies simultaneously, leading to positive and statistically significant correlations across policy types.
  - Standalone policy announcements are rare, complicating isolation of specific policy impacts.

### Fiscal policy findings
- Frequency and types:
  - Grants were the most common fiscal policy measure in both AEs and EMDEs in 2020.
  - In AEs, public loans and guarantees were the next most frequently used instruments after grants.
  - In EMDEs, tax reliefs were used mostly after grants (Figure 2, Panel C).
- Magnitude and cross-country distribution:
  - Size of fiscal measures was larger among AEs relative to EMDEs.
  - Country-level medians:
    - Median size of all fiscal measures combined in AEs: 15.2 percent of GDP.
    - Median size of all fiscal measures combined in EMDEs: 4.1 percent of GDP.
  - Dispersion:
    - Standard deviation for AEs: 7.5 percentage points.
    - Standard deviation for EMs: 5.6 percentage points.
  - Example policy: Australia government announced a $17.6 billion economic stimulus package on March 12.

### Monetary policy findings
- Instruments and usage:
  - Advanced economies (AEs) relied more on asset purchases (given policy rates at or close to effective lower bound).
  - EMDEs introduced more changes in interest rates and reserve requirements.
  - Credit and liquidity injections were widely used by both groups.
- Asset purchases:
  - Median cumulative asset purchases among AEs: 17.7 percent of GDP.
  - Standard deviation for AEs asset purchases: 6.5 percentage points.
  - Median cumulative asset purchases among EMDEs: 1.3 percent of GDP.
  - Standard deviation for EMDEs asset purchases: 2.7 percentage points.
- Policy rate cuts:
  - Median (mean) policy rate cut by AEs in 2020: 1.5 (1.1) percentage points.
    - These AEs are: Australia, Canada, Czech Republic, Hong Kong, Israel, South Korea, Norway, the United Kingdom, and the United States.
  - Median (mean) policy rate cut by EMDEs in 2020: 1.25 (1.9) percentage points.
  - Positive correlations across monetary measures indicate they were applied in tandem during the pandemic.

### Prudential policy findings
- Aggregate counts in 2020:
  - AEs announced 400 capital-related prudential measures and 348 non-capital prudential measures.
  - EMDEs announced 181 capital-related prudential measures and 202 non-capital prudential measures.
- Capital-related measures:
  - Changes in buffers and dividend restrictions were introduced, but the relaxation of other capital-related measures was more prevalent, especially among AEs.
  - Relaxation of provisioning requirements was frequently adopted among EMDEs.
- Non-capital measures:
  - AEs frequently relaxed supervisory expectations and reporting requirements.
- Correlations:
  - Different types of prudential policies were introduced simultaneously, leading to positive and significant correlations across policies.

### Conclusions (key takeaways)
- The announcement-level policy tracker for 74 countries contains 5,462 policies for 2020 and is more comprehensive and accurate for the countries in the dataset than existing trackers for that period.
- Most policy announcements occurred early in the pandemic (concentrated in March and April 2020).
- Announcements were highly correlated across policy types, making isolation of specific policy impacts challenging.
- Fiscal measures were significantly larger in AEs than in EMDEs, though dispersion within AEs and EMDEs is large.
- Central banks in EMDEs primarily cut interest rates and reserve requirements; central banks in AEs relied on asset purchase programs for monetary stimulus.
- Asset purchases were strikingly larger among AEs wherever used.
- Both AEs and EMDEs announced capital and non-capital prudential measures; buffer relaxations occurred in some AEs, other capital-related relaxations were widely implemented in both groups, provisioning requirement relaxations were more common among EMDEs, and supervisory expectations and reporting requirements were relaxed in both groups.

*Source: Annex A, wpiea2022114-print-pdf (IMF Working Paper).*

### References

### wpiea2022114-print-pdf - References

### References
- Anderson, Julia, Enrico Bergamini, Sybrand Brekelmans, Aliénor Cameron, Zsolt Darvas, Marta Domínguez Jímenez, and Catarina Midões. 2020. “The Fiscal Response to the Economic Fallout from the Coronavirus”. https://www.bruegel.org/publications/datasets/covid-national-dataset/
- Asian Development Bank. 2020. ADB Covid-19 Policy Database. https://covid19policy.adb.org/s
- Cantú, Carlos, Paolo Cavallino, Fiorella De Fiore, and James Yetman. 2021. “A Global Database on Central Banks’ Monetary Responses to Covid-19,” BIS Working Papers No 934.
- English, Bill, Kristin Forbes, and Ángel Ubide (Ed.). 2021. Monetary Policy and Central Banking in the Covid Era. Centre for Economic Policy Research.
- European Systemic Risk Board. 2020. Policy Measures in Response to the Covid-19 pandemic. https://www.esrb.europa.eu/home/search/coronavirus/html/index.en.html
- Federico, Pablo, Carlos A. Vegh, and Guillermo Vuletin. 2014. “Reserve Requirement Policy over the Business Cycle,” Working Paper 20612.
- Feyen, Erik, Tatiana Alonso Gispert, Tatsiana Kliatskova, and Davide S. Mare. 2020. “Taking Stock of the Financial Sector Policy Response to COVID-19 around the World,” Policy Research Working Paper No. 9497. World Bank, Washington, DC. http://hdl.handle.net/10986/34941
- International Monetary Fund. 2020. World Economic Outlook: The Great Lockdown. Washington, DC, April.
- International Monetary Fund. (2020). Database of Country Fiscal Measures in Response to the COVID-19 Pandemic, https://www.imf.org/en/Topics/imf-and-covid19/Fiscal-Policies-Database-in-Response-to-COVID-19
- International Monetary Fund. (2020) “Policy Response to COVID-19,” https://www.imf.org/en/Topics/imf-and-covid19/Policy-Responses-to-COVID-19
- Organisation for Economic Cooperation and Development. 2021. “An In-Depth Analysis of One Year of SME and Entrepreneurship Policy Responses to Covid-19: Lessons Learned for the Path to Recovery,” OECD SME and Entrepreneurship Papers, No. 25, OECD Publishing, Paris, https://doi.org/10.1787/6407deee-en.
- Yale Program on Financial Stability. 2020. Covid-19 Financial Response Tracker, https://som.yale.edu/centers/program-on-financial-stability/covid-19-crisis

### Appendix A: Implementing Authorities
- Scope and general rules:
  - Above-the-line measures (grants, tax reliefs, tax deferrals) are generally considered only when announced by central governments.
  - Local governments are not directly covered; extraordinary transfers from central governments to local governments are included to indirectly account for local activities.
  - In cases where regional governments play a substantial role (Belgium, Bosnia and Herzegovina, Switzerland), central and regional governments are combined.
- Treatment of different instruments:
  - Equity participations, public loans, and public guarantees can be provided by governments or state-owned entities (e.g., public banks).
  - Support from central banks or international organizations is excluded, except the European Guarantee Fund managed by the European Investment Bank because it is an explicit Covid-19 response directly financed by national governments.
  - Moratoria can be implemented by government or private sector.
- Classification rules relating to central bank involvement:
  - Monetary measures should be implemented by central banks.
  - Government purchases of equity or debt securities on secondary markets are classified as equity measures or public loans (not asset purchase programs or lending operations).
  - Credit from central banks (direct lending or indirect on-lending) is classified as a lending operation (not a public loan).
- Prudential measures:
  - Implemented by competent authorities.
  - For Basel III–related measures, only national authority actions are considered (the Basel Committee itself is not a regulator).

### Appendix B: Special Classification Issues
- A. Separation of Similar Policies
  - Within central bank lending operations, distinguish credit facilities and market liquidity measures.
  - Classification draws on intended effect: whether the stated aim mainly pertains to improving short-term market liquidity.
  - Lending operations maturing in one month (e.g., 7-day repos) are generally classified as market liquidity measures.
  - Differentiate policies by implementing authority when central bank lending intertwines with public loans:
    - On-lending is classified as a credit facility if initiated by the central bank, even if channeled through public banks.
    - Rediscounting by the central bank of loans under a public loan program is recorded as a stand-alone credit facility in addition to the public loan measure.
    - If the government establishes a fund to grant public loans and the fund is housed in the central bank, these loans are classified as public loans (not credit facilities).
  - Equity injections associated with credit-related measures are recorded separately under the “equity” variable.
    - Example: Government provides 10 million dollars to capitalize a guarantee fund that can guarantee loans up to 100 million dollars → record guarantee of 100 million dollars and equity measure of 10 million dollars.
    - Example: SPV conducts asset purchases capitalized by government → record asset purchase measure and equity measure.
  - Non-refundable portion of public loans treated as grants for most recipients; public guarantees with non-refundable clauses are not treated as grants.
    - Example: Paycheck Protection Program in the U.S. is a guarantee program with non-refundable clauses; the non-refundable part is not classified as a separate grant in the tracker.
  - Footnote on on-lending:
    - On-lending explicitly asks financial intermediaries to lend out a certain portion (typically 100 percent) of credit they receive from the central bank.
- B. European-Level Measures
  - Allocation principles for European-level measures to individual countries:
    - a) Asset purchases by the European Central Bank: allocate announced amounts to each EA member state using the Eurosystem capital key as of February 1 2020.
    - b) ECB reports actual uses of lending operations by country; no additional adjustment needed.
    - c) Prudential measures announced at EA (or EU) level are duplicated for all EA (or EU) member states.
    - d) European Guarantee Fund: recorded in two steps:
      - First, equity contribution by each national government recorded as an equity measure for each country.
      - Second, if the guarantee fund announces a guarantee to entities in one country, record a public guarantee for that country.
    - e) Next Generation EU program: recorded as a fiscal announcement for each EU member state; virtually no information was available in 2020 about spending allocation, so no size information is included for this measure.

### Appendix Table A.1: List of Countries
- AEs (numbered 1–30):
  - 1 Australia
  - 2 Austria
  - 3 Belgium
  - 4 Canada
  - 5 Czech Republic
  - 6 Denmark
  - 7 Estonia
  - 8 Finland
  - 9 France
  - 10 Germany
  - 11 Greece
  - 12 Hong Kong SAR
  - 13 Ireland
  - 14 Israel
  - 15 Italy
  - 16 Japan
  - 17 Korea, South
  - 18 Latvia
  - 19 Lithuania
  - 20 Netherlands
  - 21 Norway
  - 22 Portugal
  - 23 Singapore
  - 24 Slovak Republic
  - 25 Slovenia
  - 26 Spain
  - 27 Sweden
  - 28 Switzerland
  - 29 United Kingdom
  - 30 United States
- EMDEs (numbered 31–74):
  - 31 Argentina
  - 32 Armenia
  - 33 Bangladesh
  - 34 Bosnia and Herzegovina
  - 35 Brazil
  - 36 Bulgaria
  - 37 Chile
  - 38 China
  - 39 Colombia
  - 40 Costa Rica
  - 41 Croatia
  - 42 Egypt
  - 43 El Salvador
  - 44 Ghana
  - 45 Hungary
  - 46 India
  - 47 Indonesia
  - 48 Kazakhstan
  - 49 Kuwait
  - 50 Kyrgyzstan
  - 51 Malaysia
  - 52 Mauritius
  - 53 Mexico
  - 54 Montenegro
  - 55 Morocco
  - 56 Nepal
  - 57 Nigeria
  - 58 Oman
  - 59 Pakistan
  - 60 Panama
  - 61 Philippines
  - 62 Poland
  - 63 Qatar
  - 64 Romania
  - 65 Russia
  - 66 Rwanda
  - 67 Saudi Arabia
  - 68 South Africa
  - 69 Sri Lanka
  - 70 Thailand
  - 71 Turkey
  - 72 Ukraine
  - 73 United Arab Emirates
  - 74 Vietnam
- Note: Data reflect information obtained from public sources and thus may not reflect all official stimulus measures taken by the authorities in response to COVID-19.

### Appendix Table A.2: Policy Definitions (Policy types 1–28 with size availability)
- Fiscal
  - 1. Grants
    - Definition: Spending by the central government with near-term budgetary impact. Typical examples include transfers to firms or households, health spending, transfers to local governments, subsidies to social safety nets, and other spending (such as infrastructure spending) that directly or indirectly responds to Covid-19. If a measure has a long-term nature, record the estimated impact during 2020 and 2021. Include discretionary policy actions related to automatic stabilizers such as coverage expansions and extra funding. Do not include non-discretionary spending.
    - Size: Y
  - 2. Tax reliefs
    - Definition: Reductions in any type of taxes where the amounts covered do not need to be repaid in the future. Contributions to social security and fees paid to the government are considered taxes in the tracker.
    - Size: Y
  - 3. Tax deferrals
    - Definition: Direct or indirect deferrals of any type of taxes. Examples of indirect deferrals include (i) accelerated depreciation, (ii) suspension of penalties on late tax payments.
    - Size: Y
  - 4. Equity participations
    - Definition: Direct equity participations in private or state-owned firms, or equity investments in investment funds that provide capital to firms. Examples include (i) setting up a fund that purchases shares in the secondary market, (ii) direct capital contributions, (iii) co-investment with private investors (only include public portion).
    - Size: Y
  - 5. Public loans
    - Definition: Loans granted by the public sector, either directly from the government or from state-owned financial institutions. Excludes (i) loans from foreign governments or international organizations; (ii) explicit on-lending from the central bank channeled through banks (this would be a lending operation).
    - Size: Y
  - 6. Public guarantees
    - Definition: Guarantees granted by the public sector targeting financial activities. Distinguish credit guarantees (loans to real sector) and other guarantees (funding side of financial intermediaries).
    - Size: Y
- Monetary
  - 7. Asset purchases
    - Definition: Purchases of securities (bonds, stocks, commercial paper) in the secondary market by the central bank. The intention should not be only to improve short-term market liquidity.
    - Size: Y*
  - 8. Policy rates
    - Definition: Changes in the policy interest rate. Announcements of no change or speeches on expected rate path are not considered actual policy. If multiple interest rates exist, select most related to lending as the policy rate; include other rates under “other rates.” Once selected, do not change for consistency.
    - Size: Y
  - 9. Other rates
    - Definition: Changes in important interest rates that are not the policy rate; must have broad effects.
    - Size: N
  - 10. Reserve requirements (local currency)
    - Definition: Changes related to the reserve requirement of local currency loans (ratio changes, penalty changes, calculation changes). Only changes to the ratio have a size measure; others have missing size value. If multiple reserve ratios exist, choose the one with the broadest effects as main ratio; only main ratio has sizes.
    - Size: Y
  - 11. Reserve requirements (foreign currency)
    - Definition: Same as local currency reserve requirements but for foreign currency loans.
    - Size: Y
  - 12. Lending operations
    - Definition: Central bank lending operations, including credit facilities and market liquidity measures.
    - Size: Y*
    - 12.1. Credit facilities
      - Definition: Target creation of medium- and long-term credit in response to Covid (special lending programs: direct lending, repos, rediscounting, on-lending, changes in terms). Recipients may include non-financial institutions. Intention should not be only to improve short-term market liquidity.
      - Size: Y*
    - 12.2. Market liquidity measures
      - Definition: Short-term lending or interventions in asset markets with explicit sole intention of improving short-term market liquidity. Intention determined from stated aim and context.
      - Size: N
  - 13. Foreign exchange interventions
    - Definition: Interventions intended to influence foreign exchange markets (outright purchases/sales, non-deliverable forwards, regulatory actions). Assign “+1” to measures intended to strengthen/stabilize exchange rates, “-1” to measures intended to weaken exchange rates.
    - Size: N
  - 14. Central bank swap lines
    - Definition: Swap lines between central banks. Record for counterparty with relatively greater need for foreign exchange; if cannot determine, record for both sides.
    - Size: Y
- Prudential
  - 15. Macroprudential buffers
    - Definition: Includes countercyclical capital buffer (CCyB), capital conservation buffer (CCoB), systemic risk buffer (SyRB). Sizes are actual buffer changes; postponements of scheduled future buffer changes are recognized but coded as missing size. If different banks face different changes, choose one that affects most banks for CCyB and CCoB and take a simple average for SyRB.
    - Size: Y
  - 16. Buffer usability
    - Definition: Allowing or encouraging banks to use excessive capital-related buffers (CCyB, CCoB, SyRB) without changing minimum levels or postponing planned increases.
    - Size: N
  - 17. Capital requirements
    - Definition: Capital-related rules not belonging to the three capital buffers and buffer usability (CET1, leverage ratio, total loss-absorbing capacity, risk weighting, special accounting treatments).
    - Size: N
  - 18. Dividend restrictions
    - Definition: Banks asked to partially or fully cut dividends for capital preservation.
    - Size: N
  - 19. Special provisioning rules
    - Definition: Changes to provisioning-related rules, such as provisioning ratios or loan classification, in response to Covid-19.
    - Size: N
  - 20. Borrower-based measures
    - Definition: Prudential regulations based on borrower characteristics (debt-to-income, loan-to-value, etc.).
    - Size: N
  - 21. Supervisory expectations
    - Definition: Regulators’ expectations on supervisory issues (stress testing, compliance, accounting practices).
    - Size: N
  - 22. Lending standards
    - Definition: Changes to rules or recommended practices related to bank lending standards (firm quality assessments, loan concentration, terms of credit).
    - Size: N
  - 23. Reporting requirements
    - Definition: Changes to reporting requirements to ease banks’ regulatory burden. Additional information requests for Covid monitoring are not coded as tightening if no intention of regulatory tightening.
    - Size: N
  - 24. Liquidity requirements
    - Definition: Rules related to liquidity levels banks must maintain (changes to liquidity ratios, permission to temporarily breach liquidity ratios).
    - Size: N
  - 25. Other prudential measures
    - Definition: Non-capital prudential measures on banks not included in other policy types.
    - Size: N
- Other
  - 26. Moratoria
    - Definition: Moratoria granted by government or private sector, including (i) debt moratoria; (ii) suspension of non-debt payments (rents, insurance premia, utility fees); (iii) suspension of bankruptcy.
    - Size: N
  - 27. Non-bank financial institutions
    - Definition: All prudential measures applied to non-bank financial institutions. No distinction between tightening and loosening measures. Actions unrelated to prudential regulations are not included.
    - Size: N
  - 28. Market-based measures
    - Definition: Regulations on financial market participants or recommended actions in response to Covid (short selling rules, security issuance, reporting, etc.).
    - Size: N
- Notes:
  - Y (N) indicates that size information is (not) available for the corresponding policy.
  - Y* denotes that both announced and actual sizes are recorded. For lending operations, only the credit facility subset has announced and actual sizes.

*IMF WORKING PAPERS Tracking Economic and Financial Policies During COVID-19: An Announcement-Level Database — INTERNATIONAL MONETARY FUND*

### Appendix Table A.3 List of Variables in the Database

### Appendix Table A.3 List of Variables in the Database

### Policy information (identifiers and metadata)
- ID: Policy ID — Unique ID for each line
- country: Country name
- Ifs: IFS code
- income_group: Income group — AE or EMDE
- region: Region
- NGDP_2019: Nominal GDP — Nominal 2019 GDP in millions of local currency units
- year: Year — Year of announcement
- quarter: Quarter — Quarter of announcement
- month: Month — Month of announcement
- date: Date — Date of announcement
- description: Description — Descriptions and explanations
- link: Link — Links to sources. Some links may be broken over time, therefore we also keep the corresponding ESRB/YPFS/BIS IDs, if available, for reference, though measures may not be exactly the same as the original ones because in some cases we may modify descriptions, classification, or sizes.
- var_summary: Policy summary — Summarizing the policy type(s) by line

### Broad fiscal category (aggregates and components)
- Broad_Fiscal: Dummy for fiscal policies — Sign of the sum of all policy dummies in this category
- broad_fiscal_size: Fiscal total size — Sum of all policy sizes in millions of local currency units
- broad_fiscal_gdp: Fiscal total size (% of GDP) — Sum of all policy sizes as percent of 2019 GDP

- 1. grant
  - grant_size: Grants — In millions of local currency units
  - grant_gdp: Grant size (% of GDP) — As percent of 2019 GDP

- 2. tax_relief
  - tax_relief_size: Tax relief size
  - tax_relief_gdp: Tax relief size (% of GDP)

- 3. tax_deferral
  - tax_deferral_size: Tax deferral size
  - tax_deferral_gdp: Tax deferral size (% of GDP)

- 4. equity
  - equity_size: Equity participations — Equity size
  - equity_gdp: Equity size (% of GDP)

- 5. public_loan
  - public_loan_size: Public loans — Loan size
  - public_loan_gdp: Loan size (% of GDP)

- 6. guarantee
  - guarantee_credit: Guarantee type dummy = 1 if a guarantee applies to the lending side of financial institutions; = 0 if applies to the funding side
  - guarantee_size: Guarantee size
  - guarantee_gdp: Guarantee size (% of GDP)
  - guarantee_coverage_min: Minimum coverage — Minimum coverage ratio
  - guarantee_coverage_max: Maximum coverage — Maximum coverage ratio
  - guarantee_target: Guarantee target = 1 if a guarantee applies only to SMEs; = 2 if applies only to large firms; = 0: others

### Monetary policy
- Monetary_Policy: Dummy for monetary policies — Sign of the sum of all policy dummies in this category

- 7. APP (Asset purchases)
  - APP_size: Asset purchase size — Mix of categorical and numerical values. = -111 if announced as unlimited. = -555 if the measure is a change in the terms of a previously announced measure with no change in size.
  - APP_gdp: Asset purchase size (% of GDP) — Numerical values. = 0 if the size = -111 or -555.
  - APP_used_size: Asset purchase used size — Mix of categorical and numerical values. = -999 if the used amount is zero or de minimis.
  - APP_used_gdp: Asset purchase used size (% of GDP) — Numerical values. = 0 if the used size = -999.

- 8. pol_rate (Policy rates)
  - pol_rate_size: Cuts in policy rates — The magnitude of rate cuts in basis points. Cuts are positive and hikes are negative.

- 9. other_rate: Other rates

- 10. RR_LCU (Reserve requirements — local currency)
  - RR_LCU_size: Cuts in LCU reserve requirements — In basis points. Cuts are positive and hikes are negative. = NA if it cannot be quantified or it is not the main LCU reserve requirement ratio.

- 11. RR_FX (Reserve requirements — foreign currency)
  - RR_FX_size: Cuts in FX reserve requirements — In basis points. Cuts are positive and hikes are negative. = NA if it cannot be quantified or it is not the main FX reserve requirement ratio.

### Lending operations
- 12. lending: Lending operations

- 12.1 CF (Credit facilities)
  - CF_size: Credit facility size — Mix of categorical and numerical values. = -111 if announced as unlimited. = -555 if the measure is a change in the terms of a previously announced measure with no change in size. Note that other negative values, if any, are all numerical values.
  - CF_gdp: Credit facility size (% of GDP) — Numerical values. = 0 if the size = -111 or -555.
  - CF_used_size: Credit facility used size — Mix of categorical and numerical values. = -999 if the used amount is zero or de minimis.
  - CF_used_gdp: Credit facility used size (% of GDP) — Numerical values. = 0 if the used size = -999.

- 12.2 market_liquidity: Market liquidity measures

- 13. swap_line
  - swap_line_size: Central bank swap lines — Swap line sizes
  - swap_line_gdp: Swap line sizes (% of GDP)

- 14. FXI: Foreign exchange interventions

### Prudential policies
- Prudential: Dummy for bank prudential policies — Sign of the sum of all policy dummies in this category

- 15. pru_buffer (Macroprudential buffers)
  - pru_CCyB: CCyB size — Size of CCyB cut in basis points. Cuts are positive and hikes are negative.
  - pru_CCoB: CCoB size — Size of CCoB cut
  - pru_SyRB: SyRB size — Size of SyRB cut

- 16. pru_borrower: Borrower-based measures
- 17. pru_buffer_usability: Buffer usability
- 18. pru_dividend_distribution: Dividend restrictions
- 19. pru_lending_standards: Lending standards
- 20. pru_reporting_requirements: Reporting requirements
- 21. pru_special_provisioning: Special provisioning rules
- 22. pru_supervisory_expectations: Supervisory expectations
- 23. pru_other_capital: Capital requirements
- 24. pru_other_noncapital: Other prudential policies
- 25. pru_liquidity_requirements: Liquidity requirements

### Other policies
- Other: Dummy for other policies — Sign of the sum of all policy dummies in this category

- 26. moratorium
  - moratorium_coverage: Dummy for moratorium coverage
  - Moratorium coverage = 1: debt moratoria = 0: others
  - moratorium_duration_min: Maximum duration — Minimum months
  - moratorium_duration_max: Minimum duration — Maximum months
  - moratorium_bank: Dummy for loan moratoria = 1 if a moratorium applies to bank loans; = 0 if not

- 27. market_based_measure: Market-based measures
- 28. NBFI: Non-bank financial institutions

### Supplementary information and identifiers
- extension: Dummy for extensions = 1 if a measure is an extension of an existing measure in the database
- extension_note: Extension details — The ID of the original measure, if applicable
- rollback: Dummy for rollbacks = 1 if a measure is a rollback of an existing measure in the database
- rollback_note: Rollback details — The ID of the original measure, if applicable
- european: European level — EA or EU level
- BIS_ID: BIS ID — ID in Cantú et al.(2021), if applicable
- ESRB_ID: ESRB ID — ID in the ESRB tracker, if applicable
- YPFS_ID: YPFS ID — ID in the YPFS tracker, if applicable

### Notes on conversions and calculations
- When converting currency units and calculating sizes as percent of GDP, we use nominal GDP and end-of-period exchange rates in 2019 from the October 2021 World Economic Outlook Database.

*Tracking Economic and Financial Policies During COVID-19: An Announcement-Level Database — Appendix Table A.3 List of Variables in the Database*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022114-print-pdf.pdf_