## Effects of COVID-19 Containment Measures on Economic Activity (wpiea2020158-print-pdf)

## Source details

**Canonical URL:** [Effects of COVID-19 Containment Measures on Economic Activity (wpiea2020158-print-pdf)](https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020158-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2020/english/wpiea2020158-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2020/english/wpiea2020158-print-pdf.pdf.json)

---

### I. Objectives and key questions
- Four main goals:
  - Quantify the average economic effect—across countries and measures—of containment measures using daily, high-frequency indicators (main variable: Nitrogen Dioxide (NO2) emissions).
  - Examine whether fiscal and monetary measures have mitigated the negative effects of containment measures.
  - Identify which types of containment measures produce larger economic costs and tradeoffs with infection control.
  - Assess effects of re-openings (easing containment measures) on economic activity.

### II. Data sources and indicators
- High-frequency economic indicators (coverage and start dates as reported):
  - Nitrogen Dioxide (NO2) emissions: Air Quality Open Data Platform; coverage: 62 countries total, 57 used in analysis; starting Date 1-Jan-20; units: parts per billion (ppb).
  - Total Flights: FlightRadar24; coverage: over 200 countries, 84 used; starting Date 1-Jan-20.
  - Energy Consumption: ENTSO-E transparency platform; 35 European countries; hourly total load; starting Date 1-Jan-20.
  - Maritime Import and Export Indices: AIS-derived dataset (Cerdeiro et al.); 22 countries; starting Date 1-Jan-20.
  - Retail and Transit-station Mobility: Google Mobility Reports; 73 countries; starting Date 15-Feb-20.
- Containment measures:
  - Oxford COVID-19 Government Response Tracker (OxCGRT): eight dimensions normalized to range between 0 and 1; Stringency Index is the average of sub-indices; coverage: 151 countries/regions; starting Date 1-Jan-20.
- Fiscal and monetary policy measures:
  - IMF Policy Tracker: fiscal packages (announced/implemented) in percent of GDP; monetary policy actions (policy rate changes); coverage: 195 IMF member countries.
- COVID-19 infections and deaths:
  - Johns Hopkins University Coronavirus Resource Center dashboard; coverage begins January 22, 2020; 208 countries and regions.
- NO2-based baseline sample:
  - Balanced sample of 57 economies with at least 30 observation days after 100 confirmed cases; data cut-off date: June 15, 2020.

### III. Empirical methodology
- Main estimation framework:
  - Local projections (Jordà, 2005) to estimate dynamic cumulative effects over h = 0,..,30 days.
  - Baseline regression uses daily change in log economic indicator (NO2 in baseline) as dependent variable; includes country fixed effects, controls for infections and deaths lagged one day, temperature and humidity, country-specific linear, quadratic, and cubic time trends, and lagged changes in the dependent variable.
- Heterogeneous-response specification:
  - Smooth-transition formulation with weighting function F(zit) = exp(-γzit)/(1−exp(-γzit)), γ>0, where z is normalized country-specific characteristic (e.g., fiscal stimulus, cumulative policy rate cuts).
- Infection-response estimates:
  - Local projection where dependent variable is daily change in log infections; controls similar to baseline.
- Inference:
  - Impulse responses computed for each horizon with 95 percent confidence bands; robust standard errors clustered at country level.
- NO2-to-industrial production translation:
  - Monthly panel regression for 38 countries, January 2019–April 2020: ∆IPi,t = α + β∆NO2 + μi + εi,t; estimated β = 0.015 (one percent drop in NO2 associated with 0.015 percent decline in industrial production).

### IV. Main empirical findings
- Baseline NO2 response:
  - Containment measures significantly reduced NO2 emissions.
  - In countries with stringent measures, cumulative NO2 emissions may have been reduced by almost 99 percent 30 days after implementation relative to the underlying country-specific path without intervention.
- Translation to industrial production:
  - Using β = 0.015, containment measures imply about a 15 percent decline (month-on-month) in industrial production over the 30-day period following implementation.
- Robustness:
  - Results robust to inclusion of daily time fixed effects, mobility controls (retail or transit), leads of stringency index, using contemporaneous NO2 as control and estimating impact after one day, and exclusion of China.
  - Seasonality checks on monthly NO2 (January 2019–June 2020) indicate monthly fixed effects typically not statistically significant (except July and October).
- Impact on other high-frequency indicators (30-day cumulative effects):
  - Total flights (international + domestic): reduced by more than 99 percent.
  - Total energy consumed: declined by more than 95 percent.
  - Maritime imports and exports: reduced by around 30 percent (impact more pronounced on exports).
  - Retail and transit mobility: reduced by more than 400 percentage points (cumulative percentage points relative to baseline).
- Role of macro policy responses (as of June 15, 2020):
  - Fiscal stimulus:
    - More than 90 countries had deployed or announced fiscal measures.
    - Fiscal packages ranged from less than 1 percent of GDP to as much as 12 percent of GDP for some economies.
    - Advanced Economies average fiscal stimulus: 5 percent of GDP; Emerging Market and Developing Economies average: 2.3 percent of GDP.
    - Heterogeneous-response: containment measures had a much larger adverse impact on economic activity in countries with relatively small fiscal packages—equivalent to a 22 percent decline in industrial production. In contrast, impact not statistically different from zero in countries with large fiscal stimulus.
  - Policy interest rate cuts:
    - Policy rates reduced in 97 countries from January 2019 to date.
    - In EMDEs, more than 10 countries lowered policy rates by over 200 bps; Ukraine cut by 400 bps.
    - Heterogeneous-response: in countries with large cumulative policy rate cuts, the adverse impact of containment measures was mitigated and the impact was not statistically significant; the effect was much more adverse where monetary policy was not eased.
- Cost-effectiveness across containment measures (caveat: many measures implemented simultaneously):
  - Measures analyzed individually and jointly: (i) school closures; (ii) workplace closures; (iii) cancellation of public events; (iv) restrictions on size of gatherings; (v) closures of public transport; (vi) stay-at-home orders; (vii) restrictions on internal movement; (viii) restrictions on international travel.
  - Broad findings:
    - Workplace closures, cancellations of events, and stay-at-home orders: among the most effective in curbing infections and among the costliest economically.
    - Closures of public transport and restrictions on internal movement: costly economically but less effective in curbing infections.
    - Restrictions on international travel: least costly economically and still successful in lowering COVID-19 infections.
- Effects of easing containment measures (re-openings):
  - Sample: balanced panel of 54 countries that had peaked stringency and then lowered it; horizon: 20 days following relaxation.
  - Easing increased NO2 emissions by more than 500 log percentage points in 20 days relative to a baseline of stringent measures.
  - Translating to industrial production: roughly a 7 percent increase—sizeable but much smaller in absolute value than the decline associated with tightening containment measures.

### V. Quantitative summary statistics (selected reported values)
- NO2 emissions (log): Obs. 9,170; Mean 2.0; Min -0.9; Max 4.4; Std. Dev. 0.7.
- Total Flights (log): Obs. 29,997; Mean 3.4; Min 0.0; Max 10.8; Std. Dev. 2.1.
- Retail Mobility (%): Obs. 13,456; Mean -11.7; Min -58.6; Max 2.6; Std. Dev. 13.3.
- Transit Station Mobility (%): Obs. 13,350; Mean -12.2; Min -57.8; Max 3.3; Std. Dev. 13.5.
- Maritime Import Index (log): Obs. 2,420; Mean 4.6; Min 3.83; Max 4.9; Std. Dev. 0.12.
- Energy Consumption (log): Obs. 4,785; Mean 12.1; Min 3.62; Max 15.6; Std. Dev. 1.5.
- Confirmed Cases (log): Obs. 16,996; Mean 5.3; Min -0.9; Max 14.3; Std. Dev. 3.0.
- Stringency of Measures Index (%): Obs. 24,626; Mean 0.4; Min 0; Max 1; Std. Dev. 0.4.
- Fiscal Stimulus (% of GDP): Obs. 14,290; Mean 3.3; Min 0; Max 12.1; Std. Dev. 3.1.
- Policy rate cuts (bps): Obs. 25,377; Mean 76.2; Min 0; Max 1000; Std. Dev. 118.8.

### VI. Cumulative effects by containment measure (30 days after introduction; log percentage points)
- NO2 emissions and Confirmed Cases (Table 2 reported cumulative responses):
  - Workplace closures: NO2 emissions -580; Confirmed Cases -101
  - Closures of public transport: NO2 emissions -437; Confirmed Cases -64
  - Cancellation of events: NO2 emissions -413; Confirmed Cases -149
  - School closures: NO2 emissions -368; Confirmed Cases -88
  - Restrictions on internal movement: NO2 emissions -296; Confirmed Cases -75
  - Stay-at-home requirements: NO2 emissions -296; Confirmed Cases -100
  - Restrictions on size of gathering: NO2 emissions -293; Confirmed Cases -107
  - International travel restrictions: NO2 emissions -162; Confirmed Cases -111

### VII. Policy-relevant implications and interpretation
- Tradeoffs and policy guidance:
  - Containment measures save lives but produce large short-term economic losses; the empirical evidence quantifies large adverse effects across multiple high-frequency indicators.
  - Fiscal and monetary policy response matters:
    - Larger fiscal stimulus (percent of GDP) appears to substantially mitigate short-term economic losses associated with containment measures.
    - Larger cumulative cuts in policy interest rates similarly mitigate adverse effects; limited easing is associated with larger losses.
  - Tradeoffs among containment options:
    - The most infection-reducing measures (workplace closures, event cancellations, stay-at-home orders) are also the most economically costly.
    - Less costly measures such as international travel restrictions can still reduce infections and may be prioritized when seeking lower economic disruption.
  - Re-opening dynamics:
    - Easing restrictions produces economic recovery (NO2 and implied industrial production increases) but recovery magnitude is smaller than the loss from tightening measures, implying cautious, phased re-openings may yield limited immediate economic rebound relative to the costs of lockdowns.

### VIII. Conclusions (summary points)
- Containment measures have had very large impacts on NO2 emissions and high-frequency indicators of activity.
- Estimated decline in NO2 equivalent to about a 15 percent loss in industrial production over 30 days following containment measures (using β = 0.015).
- Containment measures reduced flights and energy consumption by over 95–99 percent in the 30-day window; maritime trade and retail/transit mobility also suffered large reductions.
- Macroeconomic policy (fiscal and monetary) deployed during the crisis mitigated short-term economic losses; lack of large fiscal packages or limited monetary easing is associated with larger losses.
- Different containment measures vary in cost-effectiveness: workplace closures, event cancellations, and stay-at-home orders are most effective against infections but most economically costly; international travel restrictions are least costly and still effective.
- Re-openings lift activity (about a 7 percent implied industrial production increase over 20 days) but less than the contraction caused by tightening measures.

*Source: wpiea2020158-print-pdf (IMF).*

### References .............................................................................................................

### Effects of COVID-19 Containment Measures on Economic Activity (wpiea2020158-print-pdf)

### I. Objectives and key questions
- Four main goals:
  - Quantify the average economic effect—across countries and measures—of containment measures using daily, high-frequency indicators (main variable: Nitrogen Dioxide (NO2) emissions).
  - Examine whether fiscal and monetary measures have mitigated the negative effects of containment measures.
  - Identify which types of containment measures produce larger economic costs and tradeoffs with infection control.
  - Assess effects of re-openings (easing containment measures) on economic activity.

### II. Data sources and indicators
- Main high-frequency economic indicators assembled:
  - Nitrogen Dioxide (NO2) emissions from the Air Quality Open Data Platform (WAQI); coverage: 62 countries total, 57 used in analysis; coverage begins January 1, 2020; units: parts per billion (ppb).
  - Flights from FlightRadar24; coverage: over 200 countries, 84 used; daily data from January 1, 2020.
  - Energy consumption from ENTSO-E transparency platform; 35 European countries; hourly total load; coverage begins January 1, 2020.
  - Maritime import and export indices from AIS-derived dataset; 22 countries; coverage begins January 1, 2020.
  - Retail and transit-station mobility from Google Mobility Reports; 73 countries; coverage begins February 15, 2020.
- Containment measures:
  - Oxford COVID-19 Government Response Tracker (OxCGRT): eight dimensions (school closures; workplace closures; public event cancellations; gathering restrictions; public transport closures; stay-at-home orders; internal movement restrictions; international travel bans). Measures normalized to range between 0 and 1; Stringency Index is the average of sub-indices; data start January 1, 2020; coverage: 151 countries/regions.
- Fiscal and monetary policy measures:
  - IMF Policy Tracker: fiscal packages (announced/implemented) in percent of GDP; monetary policy actions (policy rate changes); coverage: 195 IMF member countries.
- COVID-19 infections and deaths:
  - Johns Hopkins University Coronavirus Resource Center dashboard; coverage begins January 22, 2020; 208 countries and regions.
- Sample used for NO2-based baseline analysis:
  - Balanced sample of 57 economies with at least 30 observation days after 100 confirmed cases; data cut-off date: June 15, 2020.

### III. Empirical methodology
- Primary approach:
  - Local projections (Jordà, 2005) to estimate dynamic cumulative effects of containment measures over h = 0,..,30 days.
  - Baseline regression (equation (1)) uses daily change in log economic indicator (NO2 in baseline) as dependent variable; includes country fixed effects, controls for infections and deaths lagged one day, temperature and humidity, country-specific linear, quadratic, and cubic time trends, and lagged changes in the dependent variable.
  - Heterogeneous-response specification (equation (2)): smooth-transition formulation with weighting function F(zit) = exp(-γzit)/(1−exp(-γzit)), γ>0, where z is normalized country-specific characteristic (e.g., fiscal stimulus, cumulative policy rate cuts).
- Infection-response estimates:
  - Adapted projection (equation (4)) where dependent variable is daily change in log infections; controls similar to baseline.
- Impulse responses computed for each horizon with 95 percent confidence bands; robust standard errors clustered at country level.
- NO2-to-industrial production translation:
  - Monthly panel regression (equation (3)) for 38 countries, January 2019–April 2020: ∆IPi,t = α + β∆NO2 + μi + εi,t; estimated β = 0.015 (one percent drop in NO2 associated with 0.015 percent decline in industrial production).

### IV. Main empirical findings
- Baseline NO2 response:
  - Containment measures significantly reduced NO2 emissions.
  - In countries with stringent measures, cumulative NO2 emissions may have been reduced by almost 99 percent 30 days after implementation relative to the underlying country-specific path without intervention.
- Translation to industrial production:
  - Using β = 0.015, containment measures imply about a 15 percent decline (month-on-month) in industrial production over the 30-day period following implementation.
- Robustness checks:
  - Results robust to inclusion of daily time fixed effects, mobility controls (retail or transit), leads of stringency index, using contemporaneous NO2 as control and estimating impact after one day, and exclusion of China.
  - Seasonality checks on monthly NO2 (January 2019–June 2020) indicate monthly fixed effects typically not statistically significant (except July and October).
- Impact on other high-frequency indicators (30-day cumulative effects):
  - Total flights (international + domestic): reduced by more than 99 percent.
  - Total energy consumed: declined by more than 95 percent.
  - Maritime imports and exports: reduced by around 30 percent (impact more pronounced on exports).
  - Retail and transit mobility: reduced by more than 400 percentage points (cumulative percentage points relative to baseline).
- Role of macro policy responses:
  - Fiscal stimulus (as of June 15, 2020):
    - More than 90 countries had deployed or announced fiscal measures.
    - Fiscal packages ranged from less than 1 percent of GDP to as much as 12 percent of GDP for some economies.
    - Advanced Economies average fiscal stimulus: 5 percent of GDP; Emerging Market and Developing Economies average: 2.3 percent of GDP.
    - Heterogeneous-response results: containment measures had a much larger adverse impact on economic activity in countries with relatively small fiscal packages—equivalent to a 22 percent decline in industrial production. In contrast, impact not statistically different from zero in countries with large fiscal stimulus.
  - Policy interest rate cuts:
    - Policy rates reduced in 97 countries from January 2019 to date.
    - In EMDEs, more than 10 countries lowered policy rates by over 200 bps; Ukraine cut by 400 bps.
    - Heterogeneous-response results: in countries with large cumulative policy rate cuts, the adverse impact of containment measures was mitigated and the impact was not statistically significant; the effect was much more adverse where monetary policy was not eased.
- Cost-effectiveness across containment measures:
  - Measures analyzed individually and jointly: (i) school closures; (ii) workplace closures; (iii) cancellation of public events; (iv) restrictions on size of gatherings; (v) closures of public transport; (vi) stay-at-home orders; (vii) restrictions on internal movement; (viii) restrictions on international travel.
  - Findings (subject to caution because many measures were implemented simultaneously):
    - Workplace closures, cancellations of events, and stay-at-home orders: among the most effective in curbing infections and among the costliest economically.
    - Closures of public transport and restrictions on internal movement: costly economically but less effective in curbing infections.
    - Restrictions on international travel: least costly economically and still successful in lowering COVID-19 infections.
- Effects of easing containment measures (re-openings):
  - Sample: balanced panel of 54 countries that had peaked stringency and then lowered it.
  - Horizon: 20 days following relaxation.
  - Easing containment measures increased NO2 emissions by more than 500 log percentage points in 20 days relative to a baseline of stringent measures.
  - Translating to industrial production: roughly a 7 percent increase—sizeable but much smaller in absolute value than the decline associated with tightening containment measures.

### V. Policy-relevant implications and interpretation
- Containment measures save lives but produce large short-term economic losses; empirical evidence quantifies large adverse effects across multiple high-frequency indicators.
- Fiscal and monetary policy response matters:
  - Larger fiscal stimulus (percent of GDP) appears to substantially mitigate short-term economic losses associated with containment measures.
  - Larger cumulative cuts in policy interest rates similarly mitigate adverse effects; limited easing is associated with larger losses.
- Tradeoffs among containment options:
  - Policymakers face tradeoffs: the most infection-reducing measures (workplace closures, event cancellations, stay-at-home orders) are also the costliest economically.
  - Less costly measures such as international travel restrictions can still reduce infections and may be prioritized when seeking lower economic disruption.
- Re-opening dynamics:
  - Easing restrictions produces economic recovery (NO2 and implied industrial production increases) but recovery magnitude is smaller than the loss from tightening measures, implying cautious, phased re-openings may yield limited immediate economic rebound relative to the costs of lockdowns.

### VI. Conclusions (summary points)
- Containment measures have had very large impacts on NO2 emissions and high-frequency indicators of activity.
- Estimated decline in NO2 equivalent to about a 15 percent loss in industrial production over 30 days following containment measures.
- Containment measures reduced flights and energy consumption by over 95–99 percent in the 30-day window; maritime trade and retail/transit mobility also suffered large reductions.
- Macroeconomic policy (fiscal and monetary) deployed during the crisis mitigated short-term economic losses; lack of large fiscal packages or limited monetary easing is associated with larger losses.
- Different containment measures vary in cost-effectiveness: workplace closures, event cancellations, and stay-at-home orders are most effective against infections but most economically costly; international travel restrictions are least costly and still effective.
- Re-openings lift activity (about a 7 percent implied industrial production increase over 20 days) but less than the contraction caused by tightening measures.

*wpiea2020158-print-pdf - References .............................................................................................................*

### References

### wpiea2020158-print-pdf - References

### Key references cited
- Alesina, A.F., Furceri, D., Ostry, J.D., Papageorgiou, C. and Quinn, D.P., 2019. Structural Reforms and Elections: Evidence from a World-Wide New Dataset (No. w26720). National Bureau of Economic Research.
- Akimoto H. 2003. Global Air Quality and Pollution. Science. 302(5651), pp.1719-17191
- Auerbach, A.J. and Gorodnichenko, Y., 2013. Output Spillovers from Fiscal Policy. American Economic Review, 103(3), pp.141-46.
- Cerdeiro D., Komaromi A., Lui Y. and Saeed M. 2020. World Seaborne Trade in Real Time: A Proof of Concept for Building AIS-based Nowcasts from Scratch. IMF Working Paper, Forthcoming.
- Cherniwchan J. Economic Growth, Industrialization, and the Environment. Resource and Energy Economics. 34(4), pp. 442-467
- Chinazzi et al., 2020. The effect of travel restrictions on the spread of the 2019 novel coronavirus (COVID-19) outbreak. Science, 368, 395-400.
- Coibon O., Gorodnichenko Y., Weber M, (2020). The Cost of the COVID-19 Crisis: Lockdowns, Macroeconomic Expectations, and Consumer Spending. National Bureau of Economic Research Working Paper no.27141.
- Deb P., Furceri D., Ostry J., and Tawk N. (2020). The Effects of Containment Measures on the COVID-19 Pandemic. Covid Economics: Vetted and Real-Time Papers. 2020(19), pp.53-86.
- Eichenbaum M., Rebelo S., Trabandt M., 2020. The Macroeconomics of Epidemics. NBER Working Papers 26882, National Bureau of Economic Research, Inc.
- Granger, C.W.J. and Teräsvirta, T., 1993. Modelling Nonlinear Economic Relationships Oxford University Press. New York.
- Hsiang S. et al. The effect of large-scale anti-contagion policies on the Covid-19 pandemic. Nature https://doi.org/10.1038/s41586-020-2404-8 (2020).
- Jordà, Ò., 2005. Estimation and Inference of Impulse Responses by Local Projections. American economic review, 95(1), pp.161-182.
- Kraemer et al., 2020. The effect of human mobility and control measures on the COVID-19 epidemic in China. Science, 368, 493–497.
- Kumar S., Pranab M., 2019. A Novel GDP Prediction Technique Based on Transfer Learning Using CO2 Emission Dataset. Applied Energy, 253- 113476.
- Lin, J. T. and McElroy, M. B. 2011. Detection from Space of a Reduction in Anthropogenic Emissions of Nitrogen Oxides during the Chinese Economic Downturn. Atmospheric Chemistry and Physics.
- H. Tian et al., 2020. An investigation of transmission control measures during the first 50 days of the COVID-19 epidemic in China. Science10.1126/science.abb6105 (2020).
- Ma C., Rogers J., Zhou S., 2020. Modern Pandemics: Recession and Recovery. Working paper.
- Maloney W. and Taksin T., 2020. Determinants of social distancing and economic activity during COVID-19: A global view. Covid Economics: Vetted and Real-Time Papers. 2020(13), pp.157-177.
- Marjanovic M., Milovancevic M., Mladenovic I., 2016. Prediction of GDP Growth Rate Based on Carbon Dioxide (CO2) Emissions. Journal of CO2 Utilization, 16, 212-217.
- Ramey, V.A. and Zubairy, S., 2018. Government spending multipliers in good times and in bad: evidence from US historical data. Journal of Political Economy, 126(2), pp.850-901.
- Small C., Elvidge C. Balk D, and Montgomery M. 2011. Spatial scaling of stable night lights. Remote Sensing of Environment, 115(2), pp.269-28
- Teulings, Coen N., Nikolay Zubanov. 2014. “Is Economic Recovery a Myth? Robust Estimates of Impulse Responses.” Journal of Applied Econometrics, Vol. 29: 497-514.

### Summary statistics (Table 1)
- NO2 emissions (log): Obs. 9,170; Mean 2.0; Min -0.9; Max 4.4; Std. Dev. 0.7; Source: Air Quality Open Data Platform; Starting Date 1-Jan-20; N. of countries 62
- Total Flights (log): Obs. 29,997; Mean 3.4; Min 0.0; Max 10.8; Std. Dev. 2.1; Source: FlightRadar24; Starting Date 1-Jan-20; N. of countries 217
- Retail Mobility (%): Obs. 13,456; Mean -11.7; Min -58.6; Max 2.6; Std. Dev. 13.3; Source: Google Mobility Index; Starting Date 15-Feb-20; N. of countries 132
- Transit Station Mobility (%): Obs. 13,350; Mean -12.2; Min -57.8; Max 3.3; Std. Dev. 13.5; Source: Google Mobility Index; Starting Date 15-Feb-20; N. of countries 131
- Maritime Import Index (log): Obs. 2,420; Mean 4.6; Min 3.83; Max 4.9; Std. Dev. 0.12; Source: Cerdeiro, Komaromi, Lui and Saeed (2020); Starting Date 1-Jan-20; N. of countries 22
- Maritime Export Index (log): Obs. 2,310; Mean 4.6; Min 4.21; Max 5.1; Std. Dev. 0.12; Source: Cerdeiro, Komaromi, Lui and Saeed (2020); Starting Date 1-Jan-20; N. of countries 22
- Energy Consumption (log): Obs. 4,785; Mean 12.1; Min 3.62; Max 15.6; Std. Dev. 1.5; Source: ENTSO-E; Starting Date 1-Jan-20; N. of countries 35
- Confirmed Cases (log): Obs. 16,996; Mean 5.3; Min -0.9; Max 14.3; Std. Dev. 3.0; Source: Coronavirus Resource Center of JHU; Starting Date 21-Jan-20; N. of countries 208
- Confirmed Deaths (log): Obs. 11,379; Mean 3.2; Min -1.9; Max 11.5; Std. Dev. 2.5; Source: Coronavirus Resource Center of JHU; Starting Date 22-Jan-20; N. of countries 176
- Stringency of Measures Index (%): Obs. 24,626; Mean 0.4; Min 0; Max 1; Std. Dev. 0.4; Source: OxCGRT; Starting Date 1-Jan-20; N. of countries 158
- Fiscal Stimulus (% of GDP): Obs. 14,290; Mean 3.3; Min 0; Max 12.1; Std. Dev. 3.1; Source: IMF Policy Tracker; Starting Date 1-Jan-20; N. of countries 97
- Policy rate cuts (bps): Obs. 25,377; Mean 76.2; Min 0; Max 1000; Std. Dev. 118.8; Source: IMF Policy Tracker; Starting Date 1-Jan-20; N. of countries 172

### Cumulative effects of containment measures (Table 2)
- Table reports cumulative local projection response 30 days after introduction (log percentage points)
- NO2 emissions and Confirmed Cases, cumulative effects 30 days after introduction:
  - Workplace closures: NO2 emissions -580; Confirmed Cases -101
  - Closures of public transport: NO2 emissions -437; Confirmed Cases -64
  - Cancellation of events: NO2 emissions -413; Confirmed Cases -149
  - School closures: NO2 emissions -368; Confirmed Cases -88
  - Restrictions on internal movement: NO2 emissions -296; Confirmed Cases -75
  - Stay-at-home requirements: NO2 emissions -296; Confirmed Cases -100
  - Restrictions on size of gathering: NO2 emissions -293; Confirmed Cases -107
  - International travel restrictions: NO2 emissions -162; Confirmed Cases -111
- Note: Results denote cumulative local projection response to NO2 emissions and confirmed cases to each type of containment measure. ̕ denotes results not significant 30 days after introduction. Model estimated as ∆n_{i,t+h} = u_i + θ_h c_{i,t} + X'_{i,t} Γ_h + Σ_{ℓ=1}^L ψ_{h,ℓ} ∆n_{i,t−ℓ} + ε_{i,t+h}, where ∆n_{i,t+h} = n_{i,t+h} − n_{i,t+h−1} and n_{i,t} is the logarithm of NO2 emissions (or infections). Model estimated at each horizon h=0,1,...H, with lag structure ℓ=1,2...L; c_{i,t} is index capturing different containment and mitigation measures introduced one at a time; X is matrix of time varying control variables and country-specific linear, cubic, and quadratic time trends. Results are based on June 15 data.

### Figures and empirical design notes
- Figure 1: Evolution of NO2 emissions, selected cities. NO2 emissions shown in parts per billion (ppb). Levels smoothed with a five-day moving average. Sources: Air Quality Open Data Platform, OxCGRT Stringency Index and IMF Staff calculations.
- Figure 2: Effect of Containment Measures on Total Nitrogen Dioxide (NO2) Emissions. Impulse response functions estimated using a sample of 57 countries using daily data from the start of the outbreak; restricted to countries with a significant outbreak that has lasted at least 30 days. t = 0 is date when outbreak becomes significant (100 cases) in each country. Graph shows response and confidence bands at 90 and 95 percent. Model described as for Table 2. Results based on June 15 data. Figure displays log-difference changes; text translates these into percent changes.
- Figure 3: Robustness checks: effect of Containment Measures on NO2 Emissions (deviation from the baseline, log percentage points). Same sample and restrictions as Figure 2; confidence bands at 90 and 95 percent. Results based on June 15 data.
- Figure 4: Local projection response to indicators of economic activity (Deviation from the baseline, log percentage points). Impulse response functions estimated using a sample of 119 countries using daily data; restricted to countries with a significant outbreak that has lasted at least 30 days. t = 0 when outbreak reaches 100 cases. Graph shows response and confidence bands at 95 percent. Energy consumption results are based on May 26 data. Results based on June 15 data.
- Figure 5: Policy Responses to the COVID-19 Pandemic. Panel A: Fiscal Stimulus (in percent of GDP). Panel B: Policy Rate Cuts (in basis points). Source: IMF Policy Tracker.
- Figure 6: Interaction with Fiscal and Monetary Policy (log-differences * 100). Impulse response functions estimated using a sample of 57 countries using daily data; restricted to countries with a significant outbreak that has lasted at least 30 days. Graph shows response and confidence bands at 90 and 95 percent. Model uses interaction function F(z_{it}) = exp(−γ z_{it}) (1−exp(−γ z_{it})), γ>0. Results based on June 15 data.
- Figure 7: Effect of Easing Containment Measures on Total Nitrogen Dioxide (NO2) Emissions. Impulse response functions estimated using a sample of 54 countries using daily data from the start of easing of containment measures. Restricted to countries with a significant outbreak that has lasted at least 30 days. t = 0 is date when outbreak becomes significant (100 cases) in each country. Confidence bands at 90 and 95 percent. Results based on June 15 data.
- Annex Figures (A1–A7): Additional local projection responses, containment measures and mobility indices (Apple Mobility Indices, OxCGRT Stringency Index), and alternative specifications (unsmoothed NO2, different containment measures individually and together, confirmed COVID-19 cases responses). All impulse response estimations follow the same local projection framework and sample restrictions as described above. Results based on June 15 data unless otherwise noted.

*Source: wpiea2020158-print-pdf - References (IMF).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020158-print-pdf.pdf_