## Appendix Figure A1

## Source details

**Canonical URL:** [Appendix Figure A1](https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023025-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2023/english/wpiea2023025-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2023/english/wpiea2023025-print-pdf.pdf.json)

---

### Methodology and key empirical specifications
- Estimation frameworks: Equations (1)–(4) relate quarter-on-quarter log changes in lending or firm bank debt (∆ln(L) and ∆ln(E)) to lagged policy package indicators (P), counts of policies, bank- and firm-level characteristics (X), bank/firm fixed effects (β), country × quarter fixed effects (θ), and country controls (Country controls).  
- Policy packages: a “policy package” is a specific combination of policies observed in the data defined by a matrix of dummies equal to one for countries and periods where that package is announced. Packages are lagged by one quarter to mitigate reverse causality.  
- Counts definition: packages can alternatively be defined by the sum of the number of individual policy announcements (scaled by the number of broad policy groups). Example: for a package combining fiscal and prudential policies, the count is divided by 2; for a package combining fiscal, monetary, and prudential policies, the count is divided by 3.  
- Large-size definitions: a “large” granular policy is defined as having size in the top decile in the sample (baseline). Robustness considers top quartile, tercile, and median cutoffs. Size is defined as a percent of GDP wherever applicable; large interest rate changes are calculated by taking the top quartile of changes relative to the initial level for the country.  
- Country controls include: cumulative number of COVID-19 cases per million, sovereign bond spreads (spreads to US treasuries or regional JP Morgan Bond Indices), quarterly revisions in confidential IMF forecasts of GDP growth for the duration of the crisis (to proxy intensity of the economic shock), and de facto mobility (average percent change in transit and workplace mobility indices from Google). Sovereign spread construction: average of the absolute and percent change in the yield spread from the start of the quarter to the peak using Bloomberg data.  
- Heterogeneity: bank-level heterogeneity captured by ωb (e.g., banks with equity-to-asset ratio above/below within-country median prior to the pandemic); firm-level bank-dependence captured by ωf (indicator = 1 if firm in top quartile of fraction of bank debt to total debt within country prior to the pandemic); firm quality qf is an indicator for bottom quartile within country based on four proxies averaged over 2017–19 (interest coverage ratio, return on assets, book equity scaled by assets, distance to insolvency).

### Bank-level results — main findings
- Packages that included fiscal, monetary, and prudential policies (“all-three”) had a positive and significant impact on bank lending.  
- Relative magnitudes:
  - Loan growth was approximately 300 basis points higher per quarter in the quarter following announcements of “all-three” packages, relative to a no-policy or standalone-policy (excluding fiscal-only) counterfactual.  
  - Context: in 2019, average quarterly loan growth in the sample was 200 basis points per quarter, with a standard deviation of 400 basis points.  
  - The “all-three” package lifted loan growth by more than the pre-pandemic average and by three-quarters of the pre-pandemic standard deviation.  
- Intensity (counts) results:
  - One unit increase in intensity of the “all-three” package (i.e., three additional announcements) is associated with a 44 basis point increase in quarterly credit growth. This effect is significant at the 1 percent level across specifications.  
- Large-size packages:
  - Large-size “all-three” packages (baseline: at least one granular policy in top decile) were the most effective. Based on Column 3 of Table 3, loan growth was about 600 basis points higher in the quarter following announcements of a large “all-three” package.  
  - Relaxing the large cutoff (top quartile/tercile/median) and disaggregating shows the combination of large monetary policies and large fiscal relaxations with prudential measures was particularly successful: loan growth was 460 basis points higher in the quarter following announcements of a large package that included large monetary, large fiscal, and prudential policies.  
  - The all-three-all-large package effect is noted as almost four times the effect of a large fiscal-only package in the presented comparisons.
- Bank heterogeneity:
  - Packages combining all three types of policies had a statistically significant differential effect for low-capital banks (banks with equity-to-asset ratio below within-country median).  
  - All-three-large combinations were relatively more effective for low-capital banks: estimated loan growth 100 basis points larger for less well capitalized banks.

### Firm-level results — liquidity and allocation
- Objective: link policy packages to firms’ ability to cover expenses via higher bank borrowing (∆ln(E) measured as quarter-on-quarter change in bank debt expressed as fraction of firm’s pre-pandemic expenses in months).  
- Bank-dependent firms:
  - “All-three” packages were relatively more effective for bank-dependent firms. Greater intensity and larger size of the “all-three” package increased effectiveness for bank-dependent firms.  
  - Example magnitude: the coefficient on “L.Fiscal & monetary & prudential - Large x Bank dependent” in Column 3 suggests the “all-out” package provided enough extra liquidity for firms to survive two additional months relative to non-bank dependent firms, assuming expenses remained at pre-pandemic levels.  
  - Robustness: similar results if bank dependence defined by top tercile instead of top quartile.
- Misallocation to low-quality firms:
  - Using four distinct proxies of firm quality, the triple interaction “L.Fiscal & monetary & prudential - Large x Bank dependent x Low quality” is statistically indistinguishable from zero in all specifications.  
  - Interpretation: little evidence that additional liquidity from all-out packages differentially favored ex-ante low-quality firms on average.

### Robustness checks and supplementary findings
- Baseline results resistant to multiple robustness exercises:
  - Dropping country-bank-quarters with packages that do not contain any fiscal policies (reduces sample by less than 2 percent).  
  - Including lagged credit growth to control for base effects.  
  - Alternative definition of large monetary policies using absolute change in interest rates.  
  - Including large prudential policies using counts as proxy for size.  
  - Reweighting the sample to put equal weights on countries.  
  - Focusing on weeks with combinations of one, two, or three policy announcement types to isolate “new” announcements; Figure A1 shows frequent weekly package announcements and quarters with more weeks of all-three announcements are associated with higher credit growth.  
- Granular policy-tool prevalence in successful packages:
  - Among 28 granular policy tools, each was more prevalent in successful (large-all-three) packages than in other packages. Examples noted:
    - Fiscal: grants most common in successful packages (used in all successful packages) vs 60 percent in other packages; equity injections least prevalent.  
    - Monetary: credit facilities, asset purchases, and policy rates used frequently; FX intervention (FXI) and reserve requirements less common.  
    - Prudential: relaxation of capital requirements, supervisory expectations, and reporting requirements pervasive; changes to underwriting guidance used less.  
  - At the most granular level, large-all-three combinations were mostly unique: among 38 country-quarters with above median size of large fiscal and monetary policies, each granular combination of 28 policies occurred only once.
- Reported fit and sample sizes for key regressions (selected):
  - Bank-quarter sample for Tables 1–4, 5: 7,480 bank-quarters; 1,496 banks; 49 countries. R2 ranges reported between 0.45 and 0.71 depending on specification.  
  - Firm-quarter sample for Table 6 and quality interactions: 30,675–31,035 firm-quarters; 5,943–6,207 firms; 39 countries. R2 ≈ 0.21–0.22.

### Conclusions and policy implications
- Main conclusion: Countries that announced broad packages combining fiscal, monetary, and prudential measures—and especially those that were more intensive and larger in size—saw substantially faster loan growth following the COVID-19 shock.  
- Effect heterogeneity: the impact was larger among banks that were more constrained to lend due to low capital levels and benefitted bank-dependent firms without evidence of systematic misallocation toward ex-ante low-quality firms.  
- Quantitative highlights to guide policy calibration:
  - “All-three” packages associated with ~300 basis points higher quarterly loan growth relative to no-policy/standalone-policy counterfactual.  
  - One unit greater intensity of the all-three package (three additional announcements) associated with 44 basis points higher quarterly credit growth.  
  - Large all-three packages associated with ~600 basis points higher quarterly loan growth (baseline large cutoff) and 460 basis points in specific large-monetary + large-fiscal + prudential combinations (alternative cutoffs).  
  - All-three-large packages produced about 100 basis points larger loan growth for low-capital banks.  
  - For bank-dependent firms, large all-three packages could translate to two additional months of expenses covered by extra liquidity relative to non-bank-dependent firms.  
- Policy trade-offs: while an “all out” approach in breadth and intensity appears effective following a global shock like COVID-19, constraints in policy space (particularly for emerging and developing countries) and potential costs—such as inflationary pressures and debt sustainability concerns—mean calibration of the appropriate response requires further research.

*Source: Appendix Figure A1, wpiea2023025-print-pdf*

### Appendix Figure A1.

### Appendix Figure A1

### Methodology and key empirical specifications
- Estimation frameworks: Equations (1)–(4) relate quarter-on-quarter log changes in lending or firm bank debt (∆ln(L) and ∆ln(E)) to lagged policy package indicators (P), counts of policies, bank- and firm-level characteristics (X), bank/firm fixed effects (β), country × quarter fixed effects (θ), and country controls (Country controls).  
- Policy packages: a “policy package” is a specific combination of policies observed in the data defined by a matrix of dummies equal to one for countries and periods where that package is announced. Packages are lagged by one quarter to mitigate reverse causality.  
- Counts definition: packages can alternatively be defined by the sum of the number of individual policy announcements (scaled by the number of broad policy groups). Example: for a package combining fiscal and prudential policies, the count is divided by 2; for a package combining fiscal, monetary, and prudential policies, the count is divided by 3.  
- Large-size definitions: a “large” granular policy is defined as having size in the top decile in the sample (baseline). Robustness considers top quartile, tercile, and median cutoffs. Size is defined as a percent of GDP wherever applicable; large interest rate changes are calculated by taking the top quartile of changes relative to the initial level for the country.  
- Country controls include: cumulative number of COVID-19 cases per million, sovereign bond spreads (spreads to US treasuries or regional JP Morgan Bond Indices), quarterly revisions in confidential IMF forecasts of GDP growth for the duration of the crisis (to proxy intensity of the economic shock), and de facto mobility (average percent change in transit and workplace mobility indices from Google). Sovereign spread construction: average of the absolute and percent change in the yield spread from the start of the quarter to the peak using Bloomberg data.  
- Heterogeneity: bank-level heterogeneity captured by ωb (e.g., banks with equity-to-asset ratio above/below within-country median prior to the pandemic); firm-level bank-dependence captured by ωf (indicator = 1 if firm in top quartile of fraction of bank debt to total debt within country prior to the pandemic); firm quality qf is an indicator for bottom quartile within country based on four proxies averaged over 2017–19 (interest coverage ratio, return on assets, book equity scaled by assets, distance to insolvency).

### Bank-level results — main findings
- Packages that included fiscal, monetary, and prudential policies (“all-three”) had a positive and significant impact on bank lending.  
- Relative magnitudes (based on Column 3 of Table 1 and described results):
  - Loan growth was approximately 300 basis points higher per quarter in the quarter following announcements of “all-three” packages, relative to a no-policy or standalone-policy (excluding fiscal-only) counterfactual.  
  - Context: in 2019, average quarterly loan growth in the sample was 200 basis points per quarter, with a standard deviation of 400 basis points.  
  - The “all-three” package lifted loan growth by more than the pre-pandemic average and by three-quarters of the pre-pandemic standard deviation.  
- Intensity (counts) results (Table 2):
  - One unit increase in intensity of the “all-three” package (i.e., three additional announcements) is associated with a 44 basis point increase in quarterly credit growth (Column 3). This effect is significant at the 1 percent level across specifications.  
- Large-size packages (Table 3 and Table 4):
  - Large-size “all-three” packages (baseline: at least one granular policy in top decile) were the most effective. Based on Column 3 of Table 3, loan growth was about 600 basis points higher in the quarter following announcements of a large “all-three” package.  
  - Relaxing the large cutoff (top quartile/tercile/median) and disaggregating shows the combination of large monetary policies and large fiscal relaxations with prudential measures was particularly successful: loan growth was 460 basis points higher in the quarter following announcements of a large package that included large monetary, large fiscal, and prudential policies.  
  - The all-three-all-large package effect is noted as almost four times the effect of a large fiscal-only package in the presented comparisons.
- Bank heterogeneity (Equation (2), Table 5):
  - Packages combining all three types of policies had a statistically significant differential effect for low-capital banks (banks with equity-to-asset ratio below within-country median).  
  - All-three-large combinations were relatively more effective for low-capital banks: estimated loan growth 100 basis points larger for less well capitalized banks.

### Firm-level results — liquidity and allocation
- Objective: link policy packages to firms’ ability to cover expenses via higher bank borrowing (∆ln(E) measured as quarter-on-quarter change in bank debt expressed as fraction of firm’s pre-pandemic expenses in months).  
- Bank-dependent firms:
  - “All-three” packages were relatively more effective for bank-dependent firms (Equation (3), Table 6). Greater intensity and larger size of the “all-three” package increased effectiveness for bank-dependent firms.  
  - Example magnitude: the coefficient on “L.Fiscal & monetary & prudential - Large x Bank dependent” in Column 3 suggests the “all-out” package provided enough extra liquidity for firms to survive two additional months relative to non-bank dependent firms, assuming expenses remained at pre-pandemic levels.  
  - Robustness: similar results if bank dependence defined by top tercile instead of top quartile (Table A8).  
- Misallocation to low-quality firms (Equation (4), Table 7):
  - Using four distinct proxies of firm quality, the triple interaction “L.Fiscal & monetary & prudential - Large x Bank dependent x Low quality” is statistically indistinguishable from zero in all specifications.  
  - Interpretation: little evidence that additional liquidity from all-out packages differentially favored ex-ante low-quality firms on average.

### Robustness checks and supplementary findings
- Baseline results resistant to multiple robustness exercises:
  - Dropping country-bank-quarters with packages that do not contain any fiscal policies (reduces sample by less than 2 percent) (Table A1).  
  - Including lagged credit growth to control for base effects (Table A2).  
  - Alternative definition of large monetary policies using absolute change in interest rates (Table A3).  
  - Including large prudential policies using counts as proxy for size (Table A4).  
  - Reweighting the sample to put equal weights on countries (Table A5).  
  - Focusing on weeks with combinations of one, two, or three policy announcement types to isolate “new” announcements (Table A6); Figure A1 shows frequent weekly package announcements and quarters with more weeks of all-three announcements are associated with higher credit growth.  
- Granular policy-tool prevalence in successful packages:
  - Among 28 granular policy tools, each was more prevalent in successful (large-all-three) packages than in other packages. Examples noted:
    - Fiscal: grants most common in successful packages (used in all successful packages) vs 60 percent in other packages; equity injections least prevalent.  
    - Monetary: credit facilities, asset purchases, and policy rates used frequently; FX intervention (FXI) and reserve requirements less common.  
    - Prudential: relaxation of capital requirements, supervisory expectations, and reporting requirements pervasive; changes to underwriting guidance used less.  
  - At the most granular level, large-all-three combinations were mostly unique: among 38 country-quarters with above median size of large fiscal and monetary policies, each granular combination of 28 policies occurred only once.

### Conclusions and policy implications
- Main conclusion: Countries that announced broad packages combining fiscal, monetary, and prudential measures—and especially those that were more intensive and larger in size—saw substantially faster loan growth following the COVID-19 shock.  
- Effect heterogeneity: the impact was larger among banks that were more constrained to lend due to low capital levels and benefitted bank-dependent firms without evidence of systematic misallocation toward ex-ante low-quality firms.  
- Quantitative highlights to guide policy calibration:
  - “All-three” packages associated with ~300 basis points higher quarterly loan growth relative to no-policy/standalone-policy counterfactual.  
  - One unit greater intensity of the all-three package (three additional announcements) associated with 44 basis points higher quarterly credit growth.  
  - Large all-three packages associated with ~600 basis points higher quarterly loan growth (baseline large cutoff) and 460 basis points in specific large-monetary + large-fiscal + prudential combinations (alternative cutoffs).  
  - All-three-large packages produced about 100 basis points larger loan growth for low-capital banks.  
  - For bank-dependent firms, large all-three packages could translate to two additional months of expenses covered by extra liquidity relative to non-bank-dependent firms.  
- Policy trade-offs: while an “all out” approach in breadth and intensity appears effective following a global shock like COVID-19, constraints in policy space (particularly for emerging and developing countries) and potential costs—such as inflationary pressures and debt sustainability concerns—mean calibration of the appropriate response requires further research.

*Source: Appendix Figure A1, wpiea2023025-print-pdf*

### References

### References

### Major cited works
- Acharya, V., Steffen S. (2020). The Risk of Being a Fallen Angel and the Corporate Dash for Cash in the Midst of Covid. The Review of Corporate Finance Studies, 9(3), 430-71.
- Acosta-Henao, M., Fernández, A., Gomez-Gonzales, P., & S. Kalemli-Özcan (2022). The Covid-19 shock and firm financing: government or market? Or both?
- Alfaro, L., Chari, A., Greenland, A.N., Schott, P.K. (2020). Aggregate and Firm-Level Stock Returns during Pandemics, in Real Time, NBER Working Paper No. 26950.
- Altavilla C., Barbiero F., Boucinha M., Burlon L. (2021). The Great Lockdown: Pandemic Response Policies and Bank Lending Conditions. ECB Working Paper 2465.
- Altavilla, C., Laeven, L. Peydró, J.L. (2020). Monetary and Macroprudential Policy Complementarities: Evidence from European Credit Registers. ECB Working Paper Series No 2504.
- Altavilla, C., Ellul, A., Pagano, M., Polo, A., Vlassopoulos, T. (2021). Loan Guarantees, Bank Lending and Credit Risk Reallocation. CEPR Working Paper 16727.
- Autor, D., Cho D., Crane, L., Goldar, M. Lutz, B., Montes, J., Peterman, W., Ratner, D., Villar, D. and Yildirmaz, A. (2020). An Evaluation of the Paycheck Protection Program Using Administrative Payroll Microdata. Working Paper, MIT.
- Bachas, N., Kim, O. S., Yannelis, C. (2021). Loan Guarantees and Credit Supply. Journal of Financial Economics, 139 (3), 872–894.
- BCBS (2021). Early lessons from the Covid-19 pandemic on the Basel reforms. July 2021.
- Beck, T., Keil, J. (2021). Are Banks Catching Corona? Effects of Covid on Lending in the U.S. Mimeo.
- Bergant, K., Forbes, K. (2021). Macroprudential Policy during Covid-19: The Role of Policy Space. NBER Working Paper 29346.
- Bruno, V., Shim, I., Shin, H.S. (2017). Comparative Assessment of Macroprudential Policies. Journal of Financial Stability 28: 183-202.
- Budnik, K., Dimitrov, I., Groß, J. Jancokova, M. (2021). Policies in Support of Lending Following the Coronavirus (Covid-19) Pandemic. ECB Occasional Paper No. 2021/257.
- Caldara, Dario, Etienne Gagnon, Enrique Martínez-García, and Christopher J. Neely (2020). “Monetary Policy and Economic Performance since the Financial Crisis,” Finance and Economics Discussion Series 2020-065. Washington: Board of Governors of the Federal Reserve System.
- Capelle-Blancard, G., Desroziers, A. (2020). The Stock Market is not the Economy? Insights from the COVID-19 crisis, Covid Economics: Vetted and Real-Time Papers, CEPR.
- Casanova, C., Hardy, B., Onen, M. (2021). Covid-19 Policy Measures to Support Bank Lending. BIS Quarterly Review, September 2021.
- Cascarino, G., Gallo, R., Palazzo, F. Sette, E. (2022). Public Guarantees and Credit Additionality during the Covid-19 Pandemic. Mimeo.
- Checo, S., Chen, S. (2022). Distance to Insolvency: A New Cross-Country Dataset of Corporate Vulnerability. Mimeo.
- Chodorow-Reich, G., Darmouni, O., Luck, S., Plosser, M. (2021). Bank Liquidity Provision across the Firm Size Distribution. Journal of Financial Economics.
- Couaillier, C., Reghezza, A., Rodriguez d’Acri, C., Scopelliti, A. (2022). How to release capital requirements during a pandemic? Evidence from euro area banks. ECB Working Paper Series No 2720.
- Cox, J., Greenwald, D., Ludvigson, S. (2020). What Explains the COVID-19 Stock Market? NBER Working Paper No. 27784.
- Demirgüç-Kunt, A, Pedraza, A., Ruiz-Ortega, C. (2020). Banking Sector Performance during the COVID-19 Crisis. World Bank Policy Research Working Paper No. 9363.
- Demmou, L., Calligaris, S., Franco, G., Dlugosch, D., Adalet McGowan, M., Sakha, S. (2021). Insolvency and debt overhang following the COVID-19 outbreak: assessment of risks and policy responses. OECD Economics Department Working Paper No. 1651.
- ElFayoumi, K., Hengge, M. (2021). Capital Markets, Covid-19 and Policy Measures, IMF Working Paper No. 2021/033.
- Gambacorta, L., Murcia, A. (2017). The Impact of Macroprudential Policies and their Interaction with Monetary Policy: An Empirical Analysis Using Credit Registry Data. BIS Working Paper No 636.
- Gourinchas, P. O., Kalemli-Ozcan, S., Penciakova, V., & Sander, N. (2022). Estimating SME Failures in Real Time: An Application to the COVID-19 Crisis. Mimeo.
- Granja, J., Makridis, C., Yannelis, C., & Zwick, E. (2020). Did the Paycheck Protection Program Hit the Target? Becker Friedman Institute Working Paper 2020-52.
- Greenwald, D., Krainer, J. Paul, P. (2020). The Credit Line Channel. Federal Reserve Bank of San Francisco Working Paper 2020-2026.
- Guerini, M., Nesta, L., Ragot, X., Schiavo, S. (2020). Firm Liquidity and Solvency under the Covid-19 lockdown in France. OFCE Policy Brief (76).
- Hubbard, R. G., Strain, M. R. (2020). Has the Paycheck Protection Program Succeeded? Brookings Papers on Economic Activity, Fall 2020.
- Igan, D., Mirzaei, A., Moore T. (2021). A Shot in the Arm: Stimulus Packages and Firm Performance during COVID-19. Mimeo.
- International Monetary Fund, 2009, “From Recession to Recovery: How Soon and How Strong?,” World Economic Outlook, Chapter 3.
- Jiménez, G., Laeven, L., Martínez-Miera, D., & Peydró J.-L. (2022). Public guarantees, relationship lending and bank credit: evidence from the Covid-19 crisis.
- Kirti, D., Liu, Y., Martinez Peria, S., Mishra, P., Strasky, J. (2022). Tracking Economic Policies during COVID-19: An Announcement-Level Database. IMF. Mimeo.
- Li, L., Strahan, P. E., Zhang, S. (2020). Banks as Lenders of First Resort: Evidence from the Covid-19 Crisis. The Review of Corporate Finance Studies, 9 (3), 472–500.
- Potter , Simon M, and Frank Smets (2019), Unconventional monetary policy tools: a cross-country analysis. CGFS Papers No 63. October 2019.
- Takats, E., Temesvary, J., (2019). How Does the Interaction of Macroprudential and Monetary Policies affect Cross-Border Bank Lending. BIS Working Papers No 782.
- Valencia, F., Varghese, Richard, Yao, Weijia, Yepez, Juan, 2021. Handle with Care: Regulatory Easing in Times of Covid-19. The B.E. Journal of Macroeconomics.

### Figures and data sample notes
- Figure 1: Distribution of growth in net customer loans (QoQ, percent) at the country level for a sample of 49 countries using bank-quarter data. Panels B and C cover samples of 18 AEs and 31 EMs. Inclusion criteria: data available for at least 5 banks covering either 60 percent of assets reported in annual data or $100bn in assets. Loan growth at country-quarter is average of bank-level growth winsorized at the 5th and 95th percentiles within quarter. Percentiles may represent different countries in different quarters.
- Figure 2: Distribution of net customer loans indexed to 2019Q4 (pre-COVID-19) using same sample as Figure 1. Indexed loan levels are averages of bank-level data winsorized at the 5th and 95th percentiles within quarter. Percentiles may represent different countries in different quarters.
- Figure 3: Shows how country policy announcements were distributed into packages comprising fiscal, monetary, and prudential policies at a quarterly frequency in 2020.
- Figure 4: Panel A shows average number of policies in each package type and quarter (only observed packages shown). Panel B shows number of fiscal policies (above the line, below the line, and contingent measures) adopted by each country in the sample during 2020 together with their cumulative size as percentage of GDP.
- Figure 5 (size distribution): Distribution of policies with sizes available and indicates the top 10th percentile threshold used as a cut off to define large policies. Measurement notes: sizes for fiscal policy measures, guarantees, asset purchase programs measured relative to 2019 GDP; cuts in monetary policy interest rate measured as a fraction of their level at the end of 2019. Sizes are cumulated within policy to the country-quarter level before identifying large policies.
- Figure 5 (granular composition): Prevalence of individual granular policies across policy packages separated into two groups: large packages (components with above median sizes as in Column 4 of Table 4) with all three policy types; and all other packages. Panels cover Fiscal, Monetary, Prudential policies.

### Key regression results (tables; bank-quarter regressions unless noted)
- Table 1: Dependent variable: log change in net customer loans in basis points (QoQ ln change x 10000), winsorized at the 5th and 95th percentiles by quarter. Sample: 2020Q1-2021Q1; 49 countries; 1,496 banks; 7,480 bank-quarters.
  - Fiscal only: 423.7 *** (125.5) in column (1); 194.0 (207.0) in column (2); 169.1 (264.3) in column (3).
  - Fiscal & monetary only: 317.8 (212.6) in (1); 58.0 (264.2) in (2); 68.7 (310.5) in (3).
  - Fiscal & prudential only: 555.0 *** (114.8) in (1); 279.9 (191.6) in (2); 289.7 (271.1) in (3).
  - Fiscal & monetary & prudential: 487.8 *** (84.7) in (1); 273.6 ** (113.4) in (2); 294.6 (197.5) in (3).
  - R2: 0.45 (1), 0.46 (2), 0.47 (3).

- Table 2: Main independent variables are lagged counts of policies in each mutually exclusive package. Sample: 7,480 bank-quarters; 1,496 banks; 49 countries.
  - Fiscal only (count): 19.1 (15.8) in (1); -5.2 (14.7) in (2); -6.4 (14.1) in (3).
  - Fiscal & monetary only (count): 27.5 * (14.4) in (1); 0.8 (16.6) in (2); 0.5 (16.0) in (3).
  - Fiscal & prudential only (count): 53.9 * (30.1) in (1); 29.1 (28.6) in (2); 29.2 (29.4) in (3).
  - Fiscal & monetary & prudential (count): 46.7 *** (2.7) in (1); 41.5 *** (5.8) in (2); 43.7 *** (9.8) in (3).
  - R2: 0.48 (1), 0.51 (2), 0.52 (3).

- Table 3: Packages separated into those that contain at least one large element (top decile) and those that do not. Sample: 7,480 bank-quarters; 1,496 banks; 49 countries.
  - Fiscal only - Other: 339.8 *** (122.1) in (1); 55.0 (133.3) in (2); -50.1 (142.5) in (3).
  - Fiscal & monetary only - Other: 280.5 (205.3) in (1); -14.7 (215.2) in (2); -65.2 (220.6) in (3).
  - Fiscal & prudential only - Other: 487.1 *** (110.2) in (1); 183.2 (152.8) in (2); 94.9 (167.5) in (3).
  - Fiscal & monetary & prudential only - Other: 371.0 *** (105.5) in (1); 160.2 (110.6) in (2); 90.9 (128.0) in (3).
  - Fiscal & monetary & prudential only - Large: 888.0 *** (125.8) in (1); 722.9 *** (187.2) in (2); 693.1 *** (212.0) in (3).
  - R2: 0.49 (1), 0.51 (2), 0.52 (3).

- Table 4: Package large-element definitions varied across columns (Decile, Quartile, Tercile, Median). Sample: 7,480 bank-quarters; 1,496 banks; 49 countries.
  - Selected coefficients:
    - Fiscal only - Other (Decile): -54.0 (140.3) in (1).
    - Fiscal only - Large (Decile): -248.7 * (142.4) in (1).
    - Fiscal & monetary & prudential - Monetary or Fiscal Large (Decile): 264.5 * (141.5) in (1).
    - Fiscal & monetary & prudential - Fiscal & Monetary Large (Decile): 687.2 *** (209.6) in (1).
    - Fiscal & monetary & prudential - Fiscal & Monetary Large (Quartile): 663.2 *** (213.6) in (2).
    - Fiscal & monetary & prudential - Fiscal & Monetary Large (Tercile): 629.7 *** (212.3) in (3).
    - Fiscal & monetary & prudential - Fiscal & Monetary Large (Median): 463.0 ** (226.1) in (4).
  - R2: 0.52 (1), 0.52 (2), 0.52 (3), 0.49 (4).

- Table 5: Differential effects across banks with varying capital levels (low capital = 2019 equity to asset ratio below within-country median). Sample: 7,480 bank-quarters; 1,496 banks; 49 countries.
  - Interacted coefficients (examples):
    - L.Fiscal & monetary only x Low E/A: 79.1 * (46.6) in column (1); 6.2 (3.8) as a count interaction in column (2).
    - L.Fiscal & monetary & prudential x Low E/A: 90.3 ** (42.0) in (1); 4.9 ** (2.2) as a count interaction in (2).
    - L.Fiscal & monetary & prudential - Other x Low E/A (size specification): 84.5 * (43.4) in column (3).
    - L.Fiscal & monetary & prudential - Large x Low E/A (size specification): 100.8 ** (43.6) in column (3).
  - R2: 0.71 across specifications.

- Table 6: Firm-quarter regressions; dependent variable = additional liquidity (qoq change in bank debt scaled by 2019 expenses in months). Sample: 31,035 firm-quarters; 6,207 firms; 39 countries.
  - Fiscal only x Bank dependent: 1.7 (1.5) in column (1); 0.2 (0.5) in (2).
  - Fiscal & monetary only x Bank dependent: 0.4 (0.7) in (1); 0.1 (0.1) in (2).
  - Fiscal & prudential only x Bank dependent: 1.3 (1.0) in (1); 0.1 (0.2) in (2).
  - Fiscal & monetary & prudential x Bank dependent: 1.5 ** (0.7) in (1); 0.2 *** (0.0) in (2).
  - Fiscal & monetary & prudential - Large x Bank dependent: 2.1 * (1.0) in size specification (3).
  - R2: 0.21 across specifications.

### Notes on econometric specifications and controls
- Bank-quarter regressions generally include bank fixed effects and controls: lagged ln assets, deposit to liability ratio, equity to asset ratio, net customer loan to asset ratio, winsorized at the 5th and 95th percentiles by quarter.
- Health controls: cumulative COVID cases per million; de facto mobility controls: average of workplace and public transit indices from Google.
- Macro controls: quarterly revisions in IMF GDP forecasts; financial stress proxy: sovereign spread-based measure.
- Firm-quarter regressions control for lagged log firm-level revenue growth and include firm and country x quarter fixed effects.
- Standard errors clustered by country. Significance markers: *, **, and *** denote significance at the 10 percent, 5 percent, and 1 percent levels respectively.

*Content based solely on the "References" section and figure/table notes from wpiea2023025-print-pdf - References.*

### introduction of an additional dimension of firm-level heterogeneity: a proxy for pre-COVID firm quality. For each proxy 

### Introduction of an additional dimension of firm-level heterogeneity: a proxy for pre-COVID firm quality

### Methodology: proxies and interaction design
- Proxies of firm quality used (across columns): (i) interest coverage ratio; (ii) return on assets; (iii) book equity scaled by assets; and (iv) a distance to insolvency measure.
- For each proxy:
  - Use a dummy for firms in the bottom quartile within country based on the average indicator for 2017-19.
  - Include all interactions between packages and bank dependence shown in the third column of Table 6.
  - Include interactions between packages and the proxy of firm quality.
  - Include triple interactions between packages, bank dependence, and firm quality.
- Standard errors clustered by country; significance denoted by *, **, and *** at the 10 percent, 5 percent, and 1 percent levels respectively.

### Key regression results (selected coefficients and precision)
- Dependent variable: Additional liquidity (months of 2019 expenses).
- Coefficients for "Fiscal & monetary & prudential - Large x Bank dependent":
  - ICR column: 2.1 * (1.1)
  - ROA column: 2.2 ** (1.1)
  - E/A column: 2.3 * (1.2)
  - DI column: 1.3 (0.9)
- Coefficients for "Fiscal & monetary & prudential - Large x Low firm quality (pre-COVID)":
  - ICR: -0.3 (1.2)
  - ROA: 0.4 (1.0)
  - E/A: 0.7 (0.8)
  - DI: -0.4 (0.4)
- Coefficients for "Fiscal & monetary & prudential - Large x Bank dependent x Low firm quality (pre-COVID)":
  - ICR: -0.2 (0.7)
  - ROA: -0.5 (0.7)
  - E/A: -0.9 (0.9)
  - DI: 0.2 (0.5)

### Sample, fit, and fixed effects
- Firm FE: Y (all columns)
- Country x Quarter FE: Y (all columns)
- Controls included in all columns: Health Controls, De facto mobility Controls, Macro Controls, Financial Stress Controls, Firm Controls, Other packages and interactions.
- R2 values by column: 0.22, 0.21, 0.21, 0.21
- Observations and cross-section:
  - Firm-Quarters: 30,675; 31,035; 31,030; 29,715
  - Firm counts: 6,135; 6,207; 6,206; 5,943
  - Countries: 39 in all columns

### Related robustness tables and selected findings (excerpts)
- Figure A1: Distribution of multiple policy packages at a weekly frequency (notes describe counting countries announcing packages with more than one policy group vs one group in each week of 2020).
- Table A1 (Table 3 repeated with narrower omitted category): Log change in credit (BPS) — selected coefficients:
  - "Fiscal only - Other": 340.9 *** (125.2) in column (1)
  - "Fiscal & monetary & prudential only - Large": 895.2 *** (126.8) in column (1)
  - R2: 0.50, 0.52, 0.53 across columns; Bank-Quarters: 7,338; Banks: 1,496; Countries: 49
- Table A2 (controls for lagged credit growth): "Fiscal & monetary & prudential only - Large": 863.7 *** (126.2) in column (1); R2: 0.51, 0.53, 0.54; Bank-Quarters: 7,480
- Table A3 (large rate cuts defined in absolute terms): "Fiscal & monetary & prudential only - Large": 891.9 *** (126.8) in column (1); R2: 0.49, 0.51, 0.52; Bank-Quarters: 7,480
- Table A4 (counts of prudential policies as proxy for size): "Fiscal & monetary & prudential only - Large": 576.4 *** (78.2) in column (1); R2: 0.45, 0.46, 0.47; Bank-Quarters: 7,480
- Table A5 (sample reweighted): "Fiscal & monetary & prudential - Other": 462.3 *** (72.5) in column (1); "Fiscal & monetary & prudential - Large": 585.8 *** (127.9) in column (1); R2: 0.41, 0.43, 0.43; Bank-Quarters: 7,480
- Table A6 (weekly packages and bank credit): Weeks combining three types of policies: 303.0 *** (77.6) in column (1), and 257.4 *** (74.7) in column (3); R2 up to 0.50; Bank-Quarters: 7,480
- Table A7 (subsamples by bank capital level): "Fiscal & monetary & prudential only - Large": 693.1 *** (212.0) in full sample (column 1); R2: 0.52, 0.54, 0.51; Bank-Quarters: 7,480
- Table A8 (alternative definition of bank dependence — top tercile within country): selected coefficients for Additional liquidity (months of 2019 expenses):
  - "Fiscal & monetary & prudential x Bank dependent": 1.0 ** (0.5) in package-dummies specification (column 1) and 0.1 *** (0.0) in policy counts / package size specification (column 3)
  - "Fiscal & monetary & prudential - Large x Bank dependent": 1.3 * (0.7) in policy counts / package size specification (column 3)
  - Sample: Firm-Quarters 31,035; Firm 6,207; Countries 39; R2: 0.21 across columns

### Interpretation highlights from reported interactions
- Large packages that combine fiscal, monetary, and prudential policies are associated with positive and often statistically significant increases in additional liquidity for bank-dependent firms (e.g., coefficients around 2.1, 2.2, 2.3 in ICR/ROA/E/A columns with standard errors 1.1–1.2).
- Triple interactions with low pre-COVID firm quality (bottom quartile dummies) generally show small and statistically insignificant coefficients (examples: -0.2 (0.7); -0.5 (0.7); -0.9 (0.9)), suggesting limited differential effects by the pre-COVID firm-quality bottom quartile across these specifications.
- Robustness checks across alternative specifications (lagged dependent variable, alternate definitions of large packages, counts of prudential policies, reweighting, weekly counts) consistently show large combined packages producing the largest positive bank lending responses in BPS or additional liquidity.

*What Policy Combinations Worked? The Effect of Policy Packages on Bank Lending during COVID-19 — Working Paper No. [WP/2023/###]*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023025-print-pdf.pdf_