## The Impact of the IMF’s COVID-19 Support to Developing and Emerging Economies (wpiea2022261-print-pdf)

## Source details

**Canonical URL:** [The Impact of the IMF’s COVID-19 Support to Developing and Emerging Economies (wpiea2022261-print-pdf)](https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022261-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2022/english/wpiea2022261-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2022/english/wpiea2022261-print-pdf.pdf.json)

---

### Scope, data, and empirical approach
- Scope:
  - Estimates impact of IMF pandemic-related lending on economic performance of a comprehensive set of low-income countries (LICs) and emerging market economies (EMEs).
  - Sample: 134 IMF member countries (all developing and emerging economies excluding eight higher income economies with per capita GDP greater than US$20,000 and 13 fragile economies).
    - In the selected sample of 134 LICs and EMEs: 83 received IMF financing during the pandemic; 51 did not receive IMF financing (control group).
- IMF lending scale (2020–21):
  - For PRGT-eligible LICs: about US$18 billion in new concessional lending commitments; including GRA lending raises this to about US$24 billion.
  - For EMEs (GRA): about US$150 billion to 35 of 80 emerging market economies.
  - Total disbursements reported: approximately US$51.3 billion in 2020 and US$13.2 billion in 2021.
- High-frequency dataset and indicators:
  - Novel monthly database combining all IMF lending (commitments and disbursements) with high-frequency proxies: mobility (Google), nighttime lights (VIIRS SOL), and an Economic Activity Index built from seven country-level indicators (quarterly GDP, industrial production, real imports, real domestic credit, electricity consumption, tourism arrivals, and monthly coincident indicators).
  - Government COVID spending from IMF’s “Fiscal Monitor Database of Country Fiscal Measures in Response to the COVID-19 Pandemic”.
- Empirical model:
  - Two-way fixed effects (TWFE): y_it = α_i + τ_t + β Fund_it + γ X_it + ε_it, with Fund_it baseline = log of cumulative IMF disbursements.
  - Controls: cumulative COVID-19 cases per 1 million people, Stringency Index (0–100), reserves (months of import), World Bank and MDB commitments, income and governance group interactions, and alternative robustness specifications (e.g., disbursements Apr–Jun 2020 only).

### Main empirical findings (economic activity)
- Overall result:
  - IMF financing had a significant and positive impact on high-frequency proxies of economic activity by end-2021: mobility, nighttime luminosity, and a conventional Economic Activity Index.
- Magnitudes (key illustrative estimates preserved):
  - An illustrative 10-percent increase in IMF financing is associated with up to an 0.2 percentage point higher level of GDP in IMF-funded countries relative to unfunded countries, by the end of 2021.
  - Mobility channel:
    - A 10-percent increase in IMF financing leads to a 0.11 unit increase in the mobility index (column (2), Table 4).
    - Using mid-point elasticity 0.5, a 10-percent increase in IMF financing implies about 0.06 percentage point higher economic activity (retail space example).
    - Reported mobility regression coefficients (IMF Disbursement (log)) include: 1.180***, 1.069***, 1.400***, 1.370***, 1.460***, 1.445***, 0.670***, 0.713*** (standard errors as in tables).
    - Range: 0.07 to 0.15 percentage point change in mobility relative to baseline depending on space.
  - Nighttime lights channel:
    - IMF Disbursement (log) on Night Lights (log): 0.010*** (col 1) and 0.008** (col 2).
    - IMF Disbursement (log) on Night Lights (% change): 1.672*** (col 3) and 1.339*** (col 4).
    - Interpretation: a 10-percent increase in IMF financing is associated with almost 0.1 percent higher nighttime lights (log spec); using elasticity 0.5, implies about 0.05 percentage point higher economic activity.
  - Economic Activity Index:
    - IMF Disbursement (log) coefficients: 0.742*** (col 1), 1.013*** (col 2), 0.015*** (col 3, log), 0.019*** (col 4, log).
    - Log specifications: 10-percent increase in IMF lending associated with a 0.15–0.2 percent point increase in economic activity.
    - Back-of-envelope: magnitude in ballpark of 0.1–0.2 percent of GDP per 10-percent increase in IMF funding.
    - Cross-country illustrative: median funded country received US$277 million over 2020–21; top quartile received US$639 million. Increasing median to top-quartile financing would imply higher GDP by almost 3 percentage points over the two-year period (illustrative).
- Fiscal transmission evidence:
  - Some evidence IMF lending had a positive effect on government spending on pandemic responses.
  - Table 10 (Covid Spending (log) regressions): IMF Disbursement (log) coefficients include 0.023 (0.022), 0.040* (0.021), 0.013 (0.021), 0.042** (0.021); Covid Response Total (log) coefficients mixed and often not significant.
  - Example magnitude: 0.4 percent increase in government COVID-related spending or about 0.2 percentage point of GDP for the median country (columns 1–4 of Table 10).
  - Survey evidence: about four-fifths of LIC emergency financing requests intended to use part or all IMF disbursements to support government budget.

### Heterogeneity and lending modalities
- Income and institutional heterogeneity:
  - IMF financing effects are larger in poorer IMF members and those with weaker governance/institutional quality.
  - Examples (Economic Activity Index, Table 8): IMF Disbursement (log) x Income: Low = 0.853***; Mid = 0.543**; High = -0.613**.
  - Night Lights and Mobility show substantially larger coefficients for Low income / Low Rule of Law groups (e.g., Night Lights IMF Disbursement (log) x Income: Low = 3.452***).
  - Interpretation: effect in low-income countries about 60 percent larger than in middle-income countries (Table 8, column 1 comparison).
- Concessional (PRGT) vs non-concessional (GRA):
  - Table 9:
    - IMF Disbursement (log) x PRGT: Retail Mobility = 1.103***; Night Lights = 2.286***; Economic Activity = 0.529**.
    - IMF Disbursement (log) x GRA: Retail Mobility = 1.100***; Night Lights = 0.059 (not significant); Economic Activity = 0.514**.
  - Interpretation: PRGT (concessional) financing shows stronger nighttime-lights and overall economic activity associations; mobility and economic activity index effects similar across modalities.

### Robustness checks and alternative specifications
- Robustness exercises documented (selected):
  - Collapsing treatment to Apr–Jun 2020 disbursements and dropping countries with later disbursements yields results similar to baseline (Annex III, Tables AIII.2–AIII.4).
  - Scaling treatment as percent of GDP or percent of IMF quota yields positive and statistically significant associations in most specifications (Annex III, Tables AIII.5–AIII.10).
  - Substituting IMF disbursements with IMF commitments returns similar sign and magnitude (Annex III, Tables AIII.11–AIII.13).
  - Country-clustered standard errors increase SEs and reduce significance for some nighttime lights and economic-activity-log specs, but core mobility, level economic-activity index, and log nighttime lights results remain positive and significant (Annex III, Tables AIII.14–AIII.16).
  - Emergency-financing-only specifications preserve positive and significant impacts across outcome variables (Annex III, Tables AIII.17–AIII.19).
- Common statistical patterns:
  - Cumulative COVID-19 cases (log) generally negative for mobility and nighttime lights; Stringency Index consistently negative for mobility.
  - Reserves and WB/MDB funds often enter with significant coefficients (sometimes negative for economic activity).

### Interpretation, limitations, and policy implications
- Interpretation:
  - Evidence suggests IMF interventions, which included sizeable budgetary support during the health crisis, can be effective in supporting economies through acute crises by sustaining mobility and other real-activity proxies.
  - Estimated effects reflect mitigation of downturn (supporting activity) rather than necessarily boosting long-run growth.
- Quantitative summary:
  - Effects roughly in the range 0.05–0.2 percent of GDP per 10-percent increase in IMF funding, depending on indicator and specification.
- Caveats and open questions:
  - Results are averages over the post-financing period; persistence of short-term effects over longer horizons remains an open question.
  - Some control coefficients display unexpected signs in certain specifications—possible omitted variable bias; controlling for income, governance, and political stability modifies some coefficients.
  - Questions remain on whether IMF interventions were efficient or at the optimal scale.
- Policy-relevant implications:
  - IMF financing, including direct budget support especially via concessional PRGT facilities, is associated with measurable support to economic activity during the COVID-19 crisis, with larger marginal benefits in lower-income and more resource-constrained members.
  - PRGT capacity remains limited relative to broader development finance needs (pre-pandemic PRGT envelope about SDR 1¼ billion per year; increased in 2021 to SDR 1.65 billion per year).

*Source: IMF Working Paper excerpts from wpiea2022261-print-pdf*

### EXECUTIVE SUMMARY _______________________________________________________ 5

### EXECUTIVE SUMMARY

### Scope and data
- Study estimates impact of IMF pandemic-related lending on economic performance of a comprehensive set of low-income countries (LICs) and emerging market economies (EMEs).
- Uses a newly compiled dataset covering all IMF lending facilities and several high-frequency proxies of economic activity.
- Exploits variation in amounts of IMF financing received by countries during the pandemic.

### Main findings
- IMF financing had a significant and positive impact on key proxies of economic activity by the end of 2021, including:
  - Two indirect measures: mobility and nighttime luminosity.
  - One direct measure: a high-frequency indicator comprising aggregate demand and production.
- Magnitude: An illustrative 10-percent increase in IMF financing is associated with up to 0.2 percentage point higher level of GDP in IMF-funded countries relative to unfunded countries, by the end of 2021.
- The economic effect is strongest in the poorest IMF members that received concessional financing (LICs).

### Fiscal transmission evidence
- Some evidence that IMF lending had a positive effect on government spending on pandemic responses.
- This is consistent with observed use of IMF lending, especially by LICs, to ease fiscal financing pressures.

### Robustness and heterogeneity
- Baseline results hold across different econometric specifications and when including various control variables.
- Heterogeneity analysis indicates stronger impacts for concessional lending recipients and poorer IMF members.
- Robustness analyses (multiple specifications and timing/scaling alternatives) support baseline conclusions.

### Interpretation and open questions
- Evidence suggests IMF interventions, which included sizeable budgetary support during the health crisis, can be effective in supporting economies through acute crises.
- Remaining questions: whether IMF interventions are efficient or at the optimal scale, and whether the short-term effects estimated here persist over longer horizons.

*Source: EXECUTIVE SUMMARY, wpiea2022261-print-pdf*

### Introduction

### Introduction

### Pandemic impact and economic context
- The COVID-19 pandemic "killed more than 6 million people worldwide in the first two years" and "could cost about US$13.8 trillion in cumulative output loss globally through to 2024."
- The IMF’s World Economic Outlook estimated that "world output shrank by 3.1 percent in 2020."
- The pandemic disproportionately affected women, low-income and migrant workers, and other vulnerable groups, and had adverse distributional consequences.
- Lockdowns produced large negative effects on employment and earnings, and supply chain bottlenecks led to reduced supply of imported goods and higher prices.
- Developing and emerging economies were particularly vulnerable because of weaker health systems, a lower proportion of jobs that can be done remotely, weaker information and communication technology infrastructure, and stronger reliance on imported goods.

### IMF response and magnitude of lending
- In 2020–21, the IMF provided unprecedented-scale financial support to members during the pandemic.
- For low-income countries (LICs) eligible for PRGT concessional financing:
  - "IMF provided financing to 53 of 69 eligible LICs during 2020–21."
  - This totaled "about US$18 billion in new concessional lending commitments, about five times the pre-pandemic annual average."
  - Including regular (non-concessional) lending from the GRA raises this to "about US$24 billion."
- For emerging market economies (EMEs) eligible to borrow from the GRA:
  - "35 of 80 emerging market economies ... received support, totaling about US$150 billion, including under precautionary arrangements."
- IMF lending was used to support public health measures, mitigate economic downturn effects from containment measures and supply disruptions, and rebuild external buffers.
- Concessional IMF facilities mentioned: Rapid Credit Instrument (RCF), Extended Credit Facility (ECF), Standby Credit Facility (SCF).
- Non-concessional GRA instruments mentioned: Rapid Financing Instrument (RFI), Extended Fund Facility (EFF), Stand-By Arrangement (SBA), Flexible Credit Line (FCL), Precautionary and Liquidity Line (PLL).

### Research objective and data construction
- The paper estimates the impact of countries’ access to IMF funding on economic activity and government spending across a large sample of LICs and EMEs following the pandemic onset.
- The authors construct "a novel, monthly database that combines information on all IMF lending during the pandemic with high-frequency proxies of economic activity, along with important controls related to the severity of the pandemic and other sources of financing."
- High-frequency indicators used include traditional measures (quarterly GDP, monthly industrial production, monthly real imports) and non-traditional measures (nighttime lights, mobility based on cell phone usage).

### Empirical strategy and model specification
- Primary empirical approach: two-way fixed effects (TWFE) model with treatment variable representing IMF funding.
- Model form (as described): y_it = α_i + τ_t + β Fund_it + γ X_it + ε_it
  - y_it: indicator of economic activity (varies by specification).
  - α_i: country fixed effect.
  - τ_t: time (month) fixed effect.
  - Fund_it: treatment variable measuring amount and timing of IMF funding (baseline: log of cumulative IMF disbursements).
  - X_it: time-varying covariates including severity of the pandemic, intensity of lockdowns, reserves, and funds from other IFIs.
- Rationale for treatment definition:
  - Log of cumulative disbursements acts as timing indicator (0 before first disbursement, positive afterwards) and embodies size of treatment.
  - Allows for multiple treatments (multiple disbursements) within a country and differentiation between smaller and larger financing.
- Identification challenges addressed:
  - Country fixed effects for persistent unobserved heterogeneity.
  - Controls for pandemic progression and lockdown measures.
  - Controls for availability of additional financing (other IFIs, reserves).
  - Additional regressions controlling for group-specific time trends (income and governance groups).
  - Robustness check defining treatment as sum of disbursements during April–June 2020 and eliminating countries that received additional IMF financing thereafter (to approximate a differences-in-differences approach).

### Key findings
- Descriptive analysis indicates that "IMF-funded countries on average experienced a faster economic recovery compared to unfunded countries."
- TWFE estimates show a "statistically significant positive impact of IMF financing" on economic activity across all proxies.
- The effect is strongest in LICs; the marginal effect is "statistically and economically higher than for higher-income countries."
- Back-of-envelope calculation: "an illustrative 10-percent increase in IMF financing is associated with up to an 0.2 percentage point greater improvement in GDP levels (or smaller loss) by the end of 2021 in IMF-funded countries on average, relative to countries that did not receive (or receive less) IMF funding."
- Evidence supporting mechanism: a "weak but positive association between COVID-related government expenditures and IMF financing," consistent with IMF funds enabling governments to spend more on COVID response and thereby supporting recovery.

### Robustness, limitations, and contributions
- Baseline results robust to controls for:
  - Common time trends across countries.
  - Progression of the pandemic.
  - Lockdown measures.
  - Availability of additional financing from other IFIs and own reserves.
- Alternative measures of IMF lending produce results "of a similar sign and magnitude" to the baseline.
- Addressed potential TWFE pitfalls due to heterogeneous treatment effects and staggered timing via a targeted robustness exercise collapsing early disbursements.
- Contributions:
  1. First to analyze IMF lending impact on economic activity in LICs and EMEs during the COVID-19 pandemic, documenting IMF and other IFI support and economic performance for a comprehensive set of developing and emerging economies.
  2. Adds to literature on IMF interventions; situates findings in mixed empirical literature on IMF program impacts.
  3. Introduces high-frequency indicators (nighttime lights, mobility) alongside traditional monthly indicators to mitigate lack of frequent macroeconomic data in EMEs and LICs.

### Data summary and indicators
- Dataset: a novel monthly dataset of IMF lending, high-frequency indicators of economic activity, and covariates; regressions primarily use monthly data with some yearly covariates where monthly data are unavailable.
- Non-traditional measures leveraged:
  - Nighttime lights (used as proxy for economic and social outcomes and GDP growth).
  - Mobility data based on cell phone usage (used to assess lockdown impacts).
- Traditional measures compiled from national authorities include quarterly GDP estimates, monthly industrial production, and monthly real imports.

*Source: wpiea2022261-print-pdf - Introduction*

### Annex I describes in detail the data used in the descriptive analysis and

### Annex I describes in detail the data used in the descriptive analysis and regressions

### Sample and country coverage
- Sample size: 134 IMF member countries (all developing and emerging economies excluding eight higher income economies with per capita GDP greater than US$20,000 and 13 fragile economies).
- In the selected sample of 134 LICs and EMEs:
  - 83 received IMF financing during the pandemic.
  - 51 did not receive IMF financing (control group).
- Borrowing-characteristics breakdown:
  - 64 LICs eligible for concessional financing (PRGT-only and Blender).
  - 70 EMEs eligible only for non-concessional financing from the GRA.
- Table reference: Table AI.1 lists the full set of countries used (Annex I).

### IMF lending data and patterns
- Data sources:
  - Detailed internal IMF repositories on commitments and disbursements (aggregated to monthly frequency).
  - Exception: precautionary arrangements include only the amount effectively drawn as disbursement.
- Pandemic lending totals and frequency:
  - During 2020–2021, IMF made 106 emergency loan disbursements to LICs and EMEs through RCF/RFI instruments, and 96 disbursements under arrangements for multiyear programs.
  - Total disbursements: approximately US$51.3 billion in 2020 and US$13.2 billion in 2021.
- Emergency financing incidence:
  - Of the 83 countries that received some IMF financing support, 75 received emergency financing.
  - Higher incidence among PRGT-eligible countries: 80 percent received emergency financing.
  - Incidence among GRA countries: 34 percent received emergency financing.
- Facility and modality heterogeneity:
  - Some countries received emergency financing only; several received both emergency financing and financing under IMF-supported multiyear programs.
  - Gradual shift over time from rapid-disbursing emergency loans toward financing under multiyear IMF-supported programs.
- IMF access limits:
  - IMF increased access limits on use of resources several times during the pandemic; some increases were temporary, while PRGT overall access limits saw a permanent 45-percent increase (as discussed in IMF reforms referenced in the source).

### High-frequency outcome variables (proxies for economic activity)
- Mobility indicators:
  - Source: Google’s Mobility Dataset (Google LLC, 2022).
  - Base: median for corresponding day of week during the 5-week period from January 3 to end-February 2020.
  - Original data: daily frequency, averaged to monthly observations.
  - Prior: mobility positively correlated with business transactions and overall economic activity.
- Nighttime lights:
  - Source: IMF’s high-frequency COVID-19 Datahub; underlying Earth Observation Group VIIRS DNB dataset (Monthly Cloud-free DNB Composite aggregated to country level by summing nighttime lights within administrative boundaries).
  - Interpretation: sum of lights (SOL) measures nighttime luminosity; associated with economic activity.
- Economic activity index (conventional proxy):
  - Constructed from seven country-level indicators: quarterly GDP, monthly coincident indicators of GDP, industrial production, real imports, real domestic credit, electricity consumption, tourism arrivals.
  - Underlying data from country statistical agencies and central banks; normalized to 0–100 index.
  - Final indicator selected per country according to best available indicator; quarterly data interpolated to monthly via cubic spline.
  - For over 60 percent of sample:
    - Quarterly GDP selected for 48 countries.
    - Monthly coincident indicator for 20 countries.
    - Industrial production for 15 countries.
  - For remaining countries, selected indicator is one of real imports, real domestic credit, electricity consumption, or tourist arrivals.
  - Detailed description in Annex I (Table AI.2 and Table AI.3 for coverage).

### Fiscal response and additional control variables
- Government COVID spending data:
  - Source: IMF’s “Fiscal Monitor Database of Country Fiscal Measures in Response to the COVID-19 Pandemic”.
  - Includes above-the-line measures and total liquidity support (credit guarantees, equity injections, asset purchases, loans, debt assumptions).
  - Combined six published vintages up to October 2021; resulting 3–4 monthly frequency series; values between vintages imputed as last reported value.
- Other time-varying controls included in regressions:
  - Severity of the pandemic: cumulative COVID-19 cases per 1 million people (Our World in Data). Note: cases may underrepresent severity due to lack of testing; COVID deaths tested alternatively with similar results.
  - Stringency of national lockdowns: stringency index varying 0–100.
  - Other financing sources: international reserves (IMF IFS), World Bank commitments for development finance and COVID-related operations (World Bank Projects Website), commitments from other MDBs (CSIS).
  - For local currency unions: reserves assigned or imputed as specified in the source (WAEMU, CEMAC, ECCU choices).

### Sample summary statistics (Table 3)
- Outcome Variables:
  - Mobility (Feb 2020=0): Mean = -12; Std. Dev. = 25; Min = -82; Max = 83.
  - Nighttime lights (Million SOL1): Mean = 2.33; Std. Dev. = 8.45; Min = 0.002; Max = 136.75.
  - Economic activity index (Dec 2016=100): Mean = 103; Std. Dev. = 36; Min = 0; Max = 355.
  - Government Covid spending (Million USD): Mean = 9,473; Std. Dev. = 64,254; Min = 0; Max = 904,201.
- IMF Lending:
  - Disbursement (Million USD): Mean = 245; Std. Dev. = 783; Min = 0; Max = 8,102.
  - Commitment (Million USD): Mean = 254; Std. Dev. = 902; Min = 0; Max = 7,970.
- Control Variables:
  - World Bank commitment (Million USD): Mean = 184; Std. Dev. = 463; Min = 0; Max = 4,200.
  - MDB commitment (Million USD): Mean = 313; Std. Dev. = 751; Min = 0; Max = 5,504.
  - Covid cumulative cases (Per million people): Mean = 11,978; Std. Dev. = 27,510; Min = 0; Max = 241,219.
  - Stringency index (0–100): Mean = 34; Std. Dev. = 31; Min = 0; Max = 99.
  - Reserves (Months of import): Mean = 13; Std. Dev. = 47; Min = 0; Max = 568.
- Note: SOL = Sum of lights (radiance summed at pixel level within country boundaries).

### Descriptive analysis highlights
- Pre-trend similarity:
  - Nighttime lights behaved very similarly for funded and unfunded countries in months prior to the pandemic, supporting identification strategy (absence of different pre-trends).
  - Economic activity index shows similar pre-pandemic trends across funded and unfunded groups.
- Group comparisons (funded vs. unfunded, simple descriptive):
  - Mobility: both groups had similar reductions at onset; funded countries showed faster and more robust recovery in mobility on average (retail, transit, grocery, workplaces).
  - Nighttime lights: during the pandemic, funded countries experienced a greater increase in nighttime lights relative to unfunded countries.
  - Economic activity index: only a very small difference across groups; initial downturn less severe for IMF-funded countries and recovery appears to start earlier for funded countries.
- GDP growth comparisons:
  - Distribution of projected pandemic-period growth roughly similar across funded and unfunded groups.
  - Actual growth rates for unfunded countries appear slightly skewed with a somewhat larger fraction experiencing lower growth than funded ones.

### Regression approach and controls for heterogeneity
- Identification strategy:
  - Rely on heterogeneity in size of IMF lending rather than a simple funded/unfunded binary comparison.
  - Control for income level and quality of governance to mitigate selection bias due to cross-group heterogeneity (e.g., lower income correlated with weaker governance and different pandemic policy responses).
- Controls included in regressions:
  - Level of income, governance quality, pandemic severity (cases per million), stringency index, international reserves, World Bank and other MDB commitments.
  - Robustness checks include replacing IMF disbursements with IMF commitments.

### Baseline regression findings (overview)
- Mobility regressions (Table 4):
  - Parameter of interest (IMF disbursements) has expected sign and is statistically significant across specifications.
  - Larger IMF disbursements are associated with higher rates of mobility during the pandemic across different spaces.
  - Economic magnitude statement: in specifications with the full set of controls (columns 2, 4, 6, 8 of Table 4), a 10-percent increase in IMF financing is associated with (continued discussion appears in the next pages of the chapter beyond the provided excerpt).

*Source: Annex I and related sections from the IMF Working Paper (content unit: wpiea2022261-print-pdf - Annex I describes in detail the data used in the descriptive analysis and regressions).*

### 0.07 to 0.15 percentage point change in the mobility relative to the baseline, depending on the space. We

### wpiea2022261-print-pdf - 0.07 to 0.15 percentage point change in the mobility relative to the baseline, depending on the space. We

### Impact on mobility and short-term economic activity
- Mid-point elasticity of economic activity with respect to mobility index used: 0.5 (from Table 2).
- Mobility effects:
  - A 10-percent increase in IMF financing leads to a 0.11 unit increase in the mobility index (column (2), Table 4).
  - In retail spaces, a 10-percent increase in IMF financing translates to about 0.06 percent increase in economic activity (applying elasticity 0.5).
  - Illustrative statement: an illustrative 10-percent increase in IMF financing would be associated with about 0.06 percentage point higher economic activity in IMF-funded countries by the end of 2021, relative to unfunded countries.
- Table 4 regression highlights (dependent variables: percent change from baseline for Retail, Grocery, Transit, Workplaces):
  - IMF Disbursement (log) coefficients in columns reported: 1.180***, 1.069***, 1.400***, 1.370***, 1.460***, 1.445***, 0.670***, 0.713*** (standard errors shown in table).
  - Controls with expected signs and significance: Cumulative cases (log) negative and significant; Stringency Index negative and significant.
  - Reserves (months of import) and WB, MDB funds (% of GDP) show mixed significance across outcomes.

### Nighttime lights evidence
- Regression results (Table 5):
  - IMF Disbursement (log) on Night Lights (log): 0.010*** (column 1) and 0.008** (column 2).
  - IMF Disbursement (log) on Night Lights (% change from baseline): 1.672*** (column 3) and 1.339*** (column 4).
- Interpretation:
  - A 10-percent increase in IMF financing would be associated with almost 0.1 percent higher nighttime lights across IMF-funded countries on average over the period (log specification).
  - Using mid-point elasticity 0.5 for nighttime lights, a 10-percent increase in IMF financing implies about 0.05 percentage point higher economic activity (similar to mobility-based value).

### Economic activity index findings
- Table 6 regressions (Economic Activity Index and Economic Activity (log)):
  - IMF Disbursement (log) coefficients: 0.742*** (col 1), 1.013*** (col 2), 0.015*** (col 3, log), 0.019*** (col 4, log).
  - Controls show unexpected signs in some specifications (Cumulative cases (log), Stringency Index, WB, MDB funds), possibly due to omitted variable confounding.
- Economic significance:
  - Log specifications: a 10-percent increase in IMF lending would be associated with a 0.15–0.2 percent point increase in economic activity.
  - Back-of-envelope summary: positive effect on the real economy modest but non-trivial, with magnitude in the ballpark of 0.1–0.2 percent of GDP per 10 percent increase in IMF funding.
  - Cross-country illustrative magnitudes:
    - Median funded country received US$277 million over 2020-21.
    - Top quartile funded country received US$639 million.
    - Increasing financing of the median funded country to the level of top quartile country would imply a higher GDP by almost 3 percentage points over the two-year period.

### Cross-country heterogeneity
- By income and institutional quality (Tables 7–8):
  - Mobility (Retail):
    - IMF Disbursement (log) x Income: Low = 1.133***; Mid = 0.798**; High = 1.109***.
    - IMF Disbursement (log) x Rule of Law: Low = 1.198***; Mid = 1.405***; High = 0.435 (not significant).
  - Night Lights:
    - IMF Disbursement (log) x Income: Low = 3.452***; Mid = -0.573; High = -0.338 (columns 4–6).
    - IMF Disbursement (log) x Rule of Law: Low = 3.499***; Mid = 0.407; High = -0.233.
  - Economic Activity Index (Table 8):
    - IMF Disbursement (log) x Income: Low = 0.853***; Mid = 0.543**; High = -0.613**.
    - IMF Disbursement (log) x Gov Effective: Low = 0.777***; Mid = 0.344; High = 0.558**.
    - IMF Disbursement (log) x Rule of Law: Low = 1.274***; Mid = 0.267; High = -0.427.
- Interpretation:
  - IMF financing exerts a proportionally larger impact in countries with lower income and weaker governmental capacity and institutions.
  - The effect in low-income countries is about 60 percent larger than in middle-income countries (Table 8, column 1 comparison).

### Lending modality: PRGT (concessional) vs GRA (non-concessional)
- Table 9 results (dependent variables: Retail Mobility, Night Lights, Economic Activity):
  - IMF Disbursement (log) x PRGT: Retail Mobility = 1.103***; Night Lights = 2.286***; Economic Activity = 0.529**.
  - IMF Disbursement (log) x GRA: Retail Mobility = 1.100***; Night Lights = 0.059 (not significant); Economic Activity = 0.514**.
- Interpretation:
  - Additional PRGT financing is significantly and positively associated with higher mobility, nighttime lights, and economic activity for LICs.
  - The effect on nighttime lights is greater for PRGT than for GRA; mobility and economic activity index effects are similar across modalities.

### Possible fiscal transmission channel
- Evidence and conjecture:
  - During the pandemic a significant fraction of IMF disbursements were channeled for budget support, especially for LICs.
  - Survey of published IMF staff reports on emergency financing requests by LICs (March 2020–June 2022) indicates about four-fifths of those requests intended to use part or all IMF disbursements to support the government budget.
  - Funded countries sustained higher levels of spending in 2020–2021 relative to 2019 levels, whereas unfunded countries seemed to normalize fiscal spending already in 2021 (Figure 9 described).
  - Regression tests: when accounting for group-specific time trends (interacting time trends with income groups and institutional quality), IMF financing is positively and significantly associated with government expenditure.
  - Typical borrower calibration: a 10 percent increase in IMF financing corresponds roughly to US$30 million for the IMF typical borrower during the pandemic (footnote/context).

### Overall interpretation and caveats
- Summary judgment:
  - IMF COVID-19 lending is associated with measurable, positive effects on mobility, nighttime lights, and an economic activity index—indicative of mitigating the downturn during the COVID-19 health shock.
  - Magnitudes: roughly 0.05–0.2 percent of GDP per 10 percent increase in IMF funding, depending on the indicator and specification.
- Caveats noted in source:
  - Estimates are averages over the post-financing period; mobility index interpretation is analogous to binary ‘post’ effect in difference-in-differences with multiple periods.
  - Some control coefficients (COVID cases, Stringency Index, financing from other IFIs) show unexpected signs in certain specifications—potential omitted variable bias; controlling for income, governance, and political stability moves some coefficients to expected signs.
  - Findings are based on a unique time period with an unusually sharp initial downturn; broader applicability of back-of-envelope calculations is cautioned.
  - Effects reflect supporting economic activity (mitigating downturn) rather than necessarily boosting long-run growth; much IMF financing in the crisis period was budget support aimed at countercyclical fiscal policy.

*Italic: Source — IMF Working Paper content (as provided in the supplied excerpt).*

### 0.4 percent increase in government COVID-related spending or about 0.2 percentage point of GDP for

### 0.4 percent increase in government COVID-related spending or about 0.2 percentage point of GDP for

### Key regression findings on government COVID-related fiscal spending
- Table 10 estimates the impact of IMF Disbursement (log) on Covid Spending (log) and Covid Response Total (log).
- IMF Disbursement (log) coefficients and robust standard errors (Table 10):
  - Covid Spending (log):
    - (1): 0.023 (0.022)
    - (2): 0.040* (0.021)
    - (3): 0.013 (0.021)
    - (4): 0.042** (0.021)
  - Covid Response Total (log):
    - (5): -0.005 (0.023)
    - (6): 0.013 (0.022)
    - (7): -0.014 (0.022)
    - (8): 0.019 (0.022)
- Sample and fit statistics (each specification):
  - Observations: 3,070
  - R2: 0.779, 0.781, 0.779, 0.781, 0.781, 0.783, 0.781, 0.784
  - Adjusted R2: 0.770, 0.773, 0.770, 0.773, 0.772, 0.775, 0.772, 0.775
  - Residual Std. Error: 1.981, 1.971, 1.982, 1.971, 2.062, 2.050, 2.062, 2.047
- Statistical significance notation:
  - * p<0.1
  - ** p<0.05
  - *** p<0.01

### Interpretation and substantive magnitudes
- The authors interpret results as IMF financing helping sustain government expenditures, with effects concentrated in direct spending rather than liquidity-support measures.
- Combined analysis of COVID spending and COVID liquidity support measures finds no significant effect of IMF financing in some specifications.
- Back-of-envelope calculation: a 10-percent increase in IMF financing would be associated with up to 0.2 percentage point higher levels of economic activity in IMF-funded countries relative to unfunded ones.
- Example descriptive magnitude reported: 0.4 percent increase in government COVID-related spending or about 0.2 percentage point of GDP for the median country in the sample (columns 1–4 of Table 10).

### Robustness analyses summarized
- Addressing staggered timing of disbursements:
  - Redefine treatment to include disbursements made in April–June 2020 and drop countries receiving disbursements after June 2020. Results similar to baseline; IMF lending continues to have a significant impact on all measures of economic activity (Annex III, Tables AIII.2–III.4).
- Scale effects:
  - Using disbursement amounts as a percent of GDP: IMF lending remains significantly and positively correlated with all outcome variables (Annex III, Tables AIII.5–AIII.7).
  - Using disbursement amounts as a percent of IMF quota: in most specifications the effect remains positive and statistically significant (Annex III, Tables AIII.8–AIII.10).
- Alternative treatment measure:
  - Substituting IMF disbursements with new commitments yields baseline results robust (Annex III, Tables AIII.11–AIII.13).
- Clustered standard errors:
  - Country-clustered standard errors are generally larger and reduce statistical significance for the nighttime lights index and economic activity index (logs), but coefficients on remaining outcome variables (log of nighttime lights, level of economic activity index, and all measures of mobility) remain positive and statistically significant (Annex III, Tables AIII.14–AIII.16).
- Emergency financing only:
  - Estimating effect for emergency financing only (excluding Fund-program disbursements) finds positive and significant impacts across all outcome variables (Annex III, Tables AIII.17–AIII.19).

### Conclusions and policy-relevant implications
- Main conclusion: IMF financing had a significant and positive impact on high-frequency proxies of economic activity during the pandemic, including four mobility indicators, nighttime luminosity measures, and traditional high-frequency economic indicators.
- The evidence suggests IMF financing, including direct budget support, can help countries manage acute crises.
- Baseline findings hold when controlling for country heterogeneity (income, policy implementation capacity, institutions) and lending facility types (concessional vs. non-concessional; emergency vs. program).
- Results are robust across alternative timing definitions, treatment scaling (GDP, IMF quota), and treatment measures (disbursements vs. commitments).
- Heterogeneous effects by income: lower-income countries may have experienced a larger marginal benefit from IMF financing than higher-income IMF members.
- PRGT context and scale:
  - Pre-pandemic PRGT funding to support a self-sustaining lending envelope of SDR 1¼ billion (about US$ 1.7 billion) per year.
  - Increased in 2021 to SDR 1.65 billion (about US$2.4 billion) per year, but remains a fraction of IDA’s annual commitments.

### Data and sample notes (Annex I)
- Outcome variable definitions and units:
  - Mobility: Percent change from baseline (Jan–Feb 2020) for retail and recreation, grocery and pharmacy, transit stations, workplaces.
  - Nighttime lights: Sum of nighttime lights within national borders, cloud coverage and stray light adjusted.
  - Economic Activity Index: Index, Dec 2016=100, measured by seven high-frequency indicators including GDP, coincident indicators of GDP, and industrial production.
  - Government Covid-19 Spending: Sum of additional government fiscal spending and under-the-line liquidity support for Covid-19, measured in Millions of US$.
  - IMF Disbursement: Cumulative IMF disbursements made in 2020-21, Millions of US$.
  - IMF Commitment: Cumulative IMF commitments on new arrangements made in 2020-21, Millions of US$.
  - Control variables include COVID-19 cases (Case per million population), Stringency index (Index, 0–100), Reserves (Months of import), MDB commitments (Millions of US$).
- Data coverage and compilation:
  - Final high-frequency indicators were seasonally adjusted, trimmed of outliers, expressed in real terms, covered all months from December 2016 to June 2022, and normalized to December 2016=100.
  - Indicator frequency and type (Table AI.4):
    - Full sample indicators: 134
      - Monthly: 72
      - Quarterly: 62
    - Indicator types include coincident indicator of GDP (20), GDP (48), industrial production (15), imports (30), domestic credit (8), electricity consumption (2), tourist arrivals (11).
    - Data quality ratings: Strong 38, Medium 49, Weak 47.
- Sample lists: Annex I provides the sample of countries with income category and whether IMF loan received.

*IMF Working Paper excerpt: Impact of IMF COVID-19 Lending*

### Section III and Annex II).

### wpiea2022261-print-pdf - Section III and Annex II

### Figures and descriptive indicators
- Figure AI.1: Direct Measure of Monthly Economic Activity, Dec 2016–Jun 2022 (series shown for Funded and Unfunded; Start of COVID marked).
- Figure AII.1: Mobility at Different Economic Spaces, Jan 2020–Jan 2022 (retail, grocery, transit, workplaces series presented).
- Figure AII.2: Measures of Severity of Pandemic and Lockdowns, Jan 2020–Jan 2022 (cumulative cases and Stringency Index series presented).

### Regression overview (Annex III): Core covariates and general patterns
- Regressions use country fixed effects and month-by-country-group interactions (various specifications).
- Common covariates across tables include: IMF Disbursement (log) or scaled measures, Cumulative cases (log), Stringency Index, Reserves (months of import), WB, MDB funds (% of GDP).
- Outcomes analyzed: Mobility (Retail, Grocery, Transit, Workplaces), Nighttime Lights (log and % change from baseline), Economic Activity Index (Dec 2016 = 100) and variants (log).

### Selected results — IMF Disbursement (log) and Economic Activity Index
- Table AIII.1 (Dependent: Economic Activity Index):
  - IMF Disbursement (log): 0.361**, 0.532***, 0.601*** (columns 1–3 respectively).
  - Cumulative cases (log): -0.309, -0.467**, -0.424**.
  - Stringency Index: -0.190***, -0.188***, -0.183***.
  - Reserves (months of import): -0.745***, -0.671***, -0.942***.
  - WB, MDB funds (% of GDP): -2.570***, -2.616***, -2.704***.
  - Observations: 3,130. R2: 0.795, 0.794, 0.798. Adjusted R2: 0.787, 0.786, 0.790. Residual Std. Error: 15.416, 15.460, 15.311.

### IMF financing in April–June 2020: Mobility, Night Lights, Economic Activity
- Table AIII.2 (IMF Financing Apr–Jun 2020: Mobility):
  - IMF Disbursement (log) coefficients by outcome (selected): Retail 1.249***, 1.104***; Grocery 1.538***, 1.407***; Transit 1.671***, 1.714***; Workplaces 0.415*, 0.483*.
  - Cumulative cases (log): Retail -2.152***, -2.361***; Grocery -0.160, -0.250; Transit -1.795***, -1.746***; Workplaces -1.378***, -1.425***.
  - Stringency Index strongly negative across outcomes: e.g., -0.525***, -0.519*** (Retail) and -0.331***, -0.325*** (Workplaces).
  - Observations: 1,155 (most columns). R2 range: 0.684–0.795. Residual Std. Error range: 8.758–13.100.

- Table AIII.3 (IMF Financing Apr–Jun 2020: Night Lights):
  - IMF Disbursement (log): 0.012*** (Night Lights log, col 1), 0.008* (col 2); 3.385*** (Night Lights % change, col 3), 2.801*** (col 4).
  - Cumulative cases (log): -0.019***, -0.018***; -2.635***, -2.427***.
  - Reserves (months of import): 0.011**, 1.543* (cols with reserves).
  - WB, MDB funds (% of GDP): 0.012*, 2.036* (cols with WB, MDB).
  - Observations: 1,648. R2: 0.982 (cols 1–2), 0.721–0.723 (cols 3–4). Residual Std. Error: 0.259–41.125.

- Table AIII.4 (IMF Financing Apr–Jun 2020: Economic Activity Index):
  - IMF Disbursement (log): 0.714*** (col 1), 1.490*** (col 2).
  - Cumulative cases (log): 1.202***, 1.262***.
  - Stringency Index: 0.121***, 0.121***.
  - Reserves (months of import): -0.849*** (col 2).
  - WB, MDB funds (% of GDP): -4.345*** (col 2).
  - Observations: 1,985. R2: 0.819, 0.838. Residual Std. Error: 14.023, 13.292.

### Scaling treatment variables: Percent of GDP and Percent of Quota specifications
- Table AIII.5 (IMF Disbursement % of GDP: Mobility):
  - IMF Disbursement (% of GDP) coefficients by outcome: Retail 3.573***, 2.826***; Grocery 3.125***, 3.002***; Transit 2.968***, 2.902***; Workplaces 1.346***, 1.677***.
  - Cumulative cases (log) strongly negative across outcomes (e.g., -2.997***).
  - Stringency Index negative and significant (e.g., -0.488***).
  - Observations: 1,657. R2 range: 0.702–0.827. Residual Std. Error range: 8.338–13.426.

- Table AIII.6 (IMF Disbursement % of GDP: Night Lights):
  - IMF Disbursement (% of GDP): 0.025*** (Night Lights log, col 1), 0.016* (col 2); 2.791**, 1.155 (Night Lights % change).
  - Cumulative cases (log): -0.024***, -0.022***; -3.217***, -2.935***.
  - Reserves (months of import): 0.011***, 1.675*** (cols with reserves).
  - WB, MDB funds (% of GDP): 0.010**, 2.194*** (cols with WB, MDB).
  - Observations: 2,619. R2: 0.984 (cols 1–2), 0.714–0.717 (cols 3–4). Residual Std. Error: 0.225–34.682.

- Table AIII.7 (IMF Disbursement % of GDP: Economic Activity Index):
  - IMF Disbursement (% of GDP): 0.389 (col 1), 1.801*** (col 2).
  - WB, MDB funds (% of GDP): -2.834*** (col 2).
  - Observations: 3,130. R2: 0.802, 0.809. Residual Std. Error: 15.231, 14.952.

- Table AIII.8 (IMF Disbursement percent of quota: Mobility):
  - IMF Disbursement (percent of quota) coefficients by outcome: Retail 0.054***, 0.048***; Grocery 0.079***, 0.078***; Transit 0.062***, 0.061***; Workplaces 0.030***, 0.033***.
  - Cumulative cases (log) and Stringency Index coefficients similar patterns (negative, significant).
  - Observations: 1,612. R2 range: 0.706–0.830. Residual Std. Error range: 8.367–13.318.

- Table AIII.9 (IMF Disbursement percent of quota: Night Lights):
  - IMF Disbursement (percent of quota): 0.0002, 0.0001 (cols 1–2); 0.047**, 0.031* (cols 3–4).
  - Cumulative cases (log): -0.024***, -0.021***; -3.171***, -2.836***.
  - Reserves (months of import): 0.012***; 1.673*** (where included).
  - Observations: 2,543. R2: 0.983 (cols 1–2), 0.716–0.719 (cols 3–4).

- Table AIII.10 (IMF Disbursement percent of quota: Economic Activity Index):
  - IMF Disbursement (percent of quota): 0.012 (col 1), 0.023*** (col 2).
  - WB, MDB funds (% of GDP): -2.622*** (col 2).
  - Observations: 3,040. R2: 0.780, 0.788.

### Alternative treatment: IMF Commitments (log)
- Table AIII.11 (IMF Commitments: Mobility):
  - IMF Commitment (log) coefficients by outcome: Retail 1.031***, 0.860***; Grocery 1.084***, 1.009***; Transit 1.153***, 1.072***; Workplaces 0.252, 0.344**.
  - Cumulative cases (log) and Stringency Index: large negative coefficients for cases and negative for stringency across mobility outcomes.
  - Observations: 1,657. R2 range: 0.701–0.827.

- Table AIII.12 (IMF Commitments: Night Lights):
  - IMF Commitment (log): 0.012***, 0.010*** (Night Lights log); 2.151***, 1.794*** (Night Lights % change).
  - Cumulative cases (log): -0.025***, -0.022***; -3.297***, -2.996***.
  - Observations: 2,619. R2: 0.984 (cols 1–2), 0.717–0.719 (cols 3–4).

- Table AIII.13 (IMF Commitments: Economic Activity Index):
  - IMF Commitment (log): 0.746***, 1.023***.
  - WB, MDB funds (% of GDP): -2.774*** (col 2).
  - Observations: 3,130. R2: 0.803, 0.811.

### Robustness and alternative specifications
- Clustered standard errors (Tables AIII.14–AIII.16) generally confirm positive IMF Disbursement associations:
  - Table AIII.14 (Mobility, clustered SE): IMF Disbursement (log) coefficients: Retail 1.180***, 1.069***; Grocery 1.400***, 1.370***; Transit 1.460***, 1.445***; Workplaces 0.670**, 0.713***.
  - Table AIII.15 (Night Lights, clustered SE): IMF Disbursement (log) 0.010**, 0.008* (Night Lights log); 1.672, 1.339 (Night Lights % change).
  - Table AIII.16 (Economic Activity Index, clustered SE): IMF Disbursement (log) 0.742* (col 1), 1.013** (col 2).

- Alternative financing type — Emergency financing only (Tables AIII.17–AIII.19):
  - Table AIII.17 (Emergency Financing: Mobility): IMF Disbursement (log) coefficients include 1.003***, 1.006***, 0.883***, 0.980***, 1.591***, 1.836*** across mobility outcomes.
  - Table AIII.18 (Emergency Financing: Night Lights): IMF Disbursement (log) 0.015***, 0.012*** (Night Lights log); 3.075***, 2.181*** (Night Lights % change).
  - Table AIII.19 (Emergency Financing: Economic Activity Index):
    - IMF Disbursement (log): 0.652***, 1.461*** (Economic Activity Index specs).
    - Economic Activity (log) models: IMF Disbursement (log) 0.006**, 0.017***.
    - WB, MDB funds (% of GDP): -4.148*** in levels spec.

### Key empirical patterns and statistical notes
- IMF disbursement (variously measured as log, percent of GDP, percent of quota, or commitments) is generally positively associated with mobility measures, nighttime lights, and the Economic Activity Index across multiple specifications and robustness checks.
- Cumulative COVID-19 cases (log) are generally negatively associated with mobility and nighttime lights; mixed signs in Economic Activity Index specifications depending on model and period.
- Stringency Index is consistently negative and statistically significant for mobility outcomes; stringency effects on Night Lights and Economic Activity vary by specification.
- Reserves (months of import) and WB, MDB funds (% of GDP) frequently enter with statistically significant coefficients (often negative for WB, MDB funds on the Economic Activity Index).
- Statistical significance notation used in tables: * p<0.1, ** p<0.05, *** p<0.01.
- Observation counts, R2, Adjusted R2, and Residual Std. Error values are reported for each model in the tables.

*IMF Working Papers — Impact of IMF COVID-19 Lending (Section III and Annex II; tables and figures as provided).*

### 838. Tokyo: Asian Development Bank Institute.

### The Impact of the IMF’s COVID-19 Support to Developing and Emerging Economies — References and Source Materials (excerpt)

### Data sources and empirical proxies cited
- Earth Observation Group. 2022. Nighttime Light, VIIRS. Available online at https://eogdata.mines.edu/products/vnl/. Accessed: 1 July 2022.
- Elvidge, C. D., K. Baugh, M. Zhizhin, F. C. Hsu, and T. Ghosh. 2017. “VIIRS Night-time Lights,” International Journal of Remote Sensing, Vol. 38, pp. 5860–5879.
- Google LLC. 2022. "Google COVID-19 Community Mobility Reports". Available online at https://www.google.com/covid19/mobility/ Accessed: <date>.
- Gibson, J., S. Olivia, and G. Boe-Gibson. 2021. “Night Lights in Economics: Sources and Uses.” Journal of Economic Surveys, Vol. 34, Issue 5, pp. 955–980.
- Roberts, M. 2021. “Tracking Economic Activity in Response to the COVID-19 Crisis using Nighttime Lights – the Case of Morocco.” Development Engineering, No. 6, pp. 1–13.
- Haldane, A., and S. Chowla. 2021. Fast Economic Indicators. Nature Reviews Physics, 3, 68-69.
- Woloszko, N. 2020. “Tracking Activity in Real Time with Google Trends.” Economics Department Working Paper No. 1634. OECD: Paris.
- Gamtkitsulashvili, T., and A. Plekhanov. 2021. “Mobility and Economic Activity Around the World During the Covid-19 Crisis.” EBRD Working Paper No. 259.

### Methodological and econometric literature referenced
- de Chaisemartin, C. and X. D’Haultfoeuille. 2020. “Two-way fixed effects estimators with heterogeneous treatment effects.” American Economic Review, 110, 2964-96.
- de Chaisemartin, C., and X. D’Haultfoeuille. 2022. “Two-Way Fixed Effects and Differences-in-Differences with Heterogeneous Treatment Effects: A Survey.” NBER Working Paper Series, No. 29691.
- Newiak, M., and T. Willems. 2017. “Evaluating the Impact of Non-Financial IMF Programs Using the Synthetic Control Method.” IMF Working Paper WP/17/109.
- Mody, A., and D. Saravia. 2003. “Catalyzing Capital Flows: Do IMF-Supported Programs Work as Commitment Devices?” IMF Working Paper WP/03/100.
- Papi, L., A. F. Presbitero, and A. Zazzaro. 2015. “IMF Lending and Banking Crises.” IMF Economic Review, Vol. 63, No. 3.
- Tirole, J. 2002. Financial Crises, Liquidity, and the International Monetary System. Princeton University Press.
- Kaufmann, D., A. Kraay, and M. Mastruzzi. 2011. “The Worldwide Governance Indicators: Methodology and Analytical Issues”. Hague Journal on the Rule of Law, Vol. 3, Issue 2, pp. 220–246, September 2011.

### Literature on IMF programs, crises, and distributional effects
- Dreher, A., and S. Walter. 2010. “Does the IMF Help or Hurt? The Effect of IMF Programs on the Likelihood and Outcome of Currency Crises.” World Development, Vol. 38, No. 1, pp. 1–18.
- Gunduz, Y. B. 2016. “The Economic Impact of Short-Term IMF Engagement in Low-Income Countries.” World Development, Vol. 87, pp. 30-49.
- Hajro, Z., and J. P. Joyce. 2009. “A True Test: Do IMF Programs Hurt the Poor? Applied Economics, Vol 41, No. 3, pp. 295–306.
- Honda, J. 2008. “Do IMF Programs Improve Economic Governance?” IMF Working Paper WP/08/114.
- Oberdabernig, D. A. 2013. “Revisiting the Effects of IMF Programs on Poverty and Inequality.” World Development, Vol. 46, pp. 113–142.
- Zettelmeyer, J. 2000. “Can Official Crisis Lending Be Counterproductive in the Short-Run?” Economic Notes, Vol. 29, No. 1, pp. 13–29.
- Independent Evaluation Office of the IMF. 2021. Growth and Adjustment in IMF-Supported Programs. Evaluation Report.

### COVID-19 economic impact, nowcasting, and related studies
- Furceri, D., M. Ganslmeier, J. D. Ostry, and N. Yang. 2021. “Initial Output Losses from the COVID-19 Pandemic: Robust Determinants.” IMF Working Paper WP/21/18.
- Miguel, E., and A. M. Mobarak. 2021. “The Economics of the COVID-19 Pandemic in Poor Countries.” NBER Working Paper Series, No. 29339.
- Sampi, J., and C. Jooste. 2020. “Nowcasting Economic Activity in Times of COVID-10: An Approximation from the Google Community Mobility Report.” Policy Research Working Paper 9247. World Bank: Washington D.C.
- Goolsbee, A., and C. Syverson. 2021. “Fear, Lockdown, and Diversion: Comparing Drivers of Pandemic Economic Decline.” Journal of Public Economics, Vol. 193, Article 104311.
- Sears, J., J. M. Villas-Boas, V. Villas-Boas, and S. B. Villas-Boas. 2021. “Are we #Stayinghome to Flatten the Curve?” American Journal of Health Economics, forthcoming.
- Molina, L., A. Jonelis, and M. Cangul. 2017. “The Informal Economy in Sub-Saharan Africa: Size and Determinants.” IMF Working Paper WP/17/156.

### Project-specific working papers and policy documents cited
- International Monetary Fund. 2021a. “Fund Concessional Financial Support for Low-Income Countries—Responding to the Pandemic.” IMF Policy Paper, July 2021.
- International Monetary Fund. 2021b. “Review of the Temporary Modifications to the Fund’s Access Limits in Response to the COVID-19 Pandemic.” IMF Policy Paper, December 2021.
- International Monetary Fund. 2022. World Economic Outlook Update: Rising Caseloads, a Disrupted Recovery, and Higher Inflation. Washington D.C., January 2022.
- Loko, B., N. Nembot, and M. Poplawski-Ribeiro. 2022. “Private Savings and COVID-19 in Sub-Saharan Africa.” IMF Working Paper WP/22/176.
- Furceri et al. 2021. IMF Working Paper WP/21/18 (see above).

### Themes indicated by the bibliography (areas of substantive analysis in the main paper)
- Use of satellite-based nightlights (VIIRS) and mobility data (Google) to track economic activity during COVID-19.
- Evaluation of the impact of IMF lending and non-financial programs on crisis outcomes, poverty, inequality, and governance.
- Econometric methods for heterogeneous treatment effects, two-way fixed effects, differences-in-differences, and synthetic control methods.
- Short-run and distributional consequences of IMF engagement in low-income and emerging economies during the COVID-19 pandemic.

*The Impact of the IMF’s COVID-19 Support to Developing and Emerging Economies. Working Paper No. WP/2022/261*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022261-print-pdf.pdf_