## 1. Overview

## Source details

**Canonical URL:** [1. Overview](https://www.imf.org/-/media/files/publications/reo/apd/2020/oct/english/text.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/reo/apd/2020/oct/english/text.pdf.md)
- [Structured JSON version](/-/media/files/publications/reo/apd/2020/oct/english/text.pdf.json)

---

### Global context and headline projections
- The global economy is beginning to recover after a sharp contraction in the second quarter of 2020 as nationwide lockdowns are lifted and replaced with more targeted containment measures.
- Global growth projected at −4.4 percent in 2020, revised up since the June 2020 World Economic Outlook (WEO) Update because of better-than-expected second quarter outturns in some major countries.
- Global growth in 2021 projected at 5.2 percent, “consistent with the expectation that social distancing persists into 2021 and fades thereafter.”

### Asia and Pacific: multispeed recovery and headline projections
- Asia and Pacific expected to contract by −2.2 percent in 2020 and to grow by 6.9 percent in 2021.
- These projections are “0.6 percentage point lower and 0.3 percentage point higher, respectively, than in the June 2020 World Economic Outlook Update.”
- Outlook varies by:
  - infection rates and containment measures;
  - scale and effectiveness of the policy response;
  - reliance on contact-intensive activities;
  - reliance on external demand.

### Recent developments and country differences
- Pandemic stages differ across Asia: some countries contained the first wave; others (Australia, Japan, Myanmar) have experienced second waves; some still strive to flatten the curve (India, Indonesia, Philippines); most Pacific island countries remain largely free of COVID-19.
- China’s activity trough was in February 2020; recovery boosted by infrastructure, real estate investment, and a surge in exports (medical/protective equipment and work-from-home electronics), followed by a gradual recovery in private nonhousing investment and consumption.
- India: activity plunged by 24 percent year-on-year in the second quarter of 2020.
- High-frequency indicators point to a trough in activity for much of Asia in April 2020, with economies recovering thereafter at multiple speeds.
- Advanced economies with lower infection rates showed a bigger pickup in activity through to August 2020 than emerging market and developing economies (excluding China).
- Inflation across Asia largely contained due to demand drop, lower oil prices, and stable food prices; inflation high in India due to supply-side disruptions related to lockdowns.
- Faster-recovering countries typically introduced effective containment measures early and timed exits well; comprehensive testing and contact tracing infrastructure were key; fiscal support facilitated resumption of activity.

### Risks, uncertainty, and scarring
- Forecasts highly uncertain; significant downside risks include:
  - a resurgence of the pandemic;
  - geopolitical tensions—particularly US-China—that may derail the recovery;
  - a rise in social unrest from the pandemic’s disproportionate impact on the poorest and most vulnerable;
  - a return to risk aversion in financial markets that could add to balance sheet vulnerabilities.
- An early, large-scale rollout of an effective vaccine is an upside risk.
- Scarring likely: labor market participation has fallen; output expected to remain below pre-pandemic trends over the medium term; the most vulnerable likely to be hit hardest.

### Policy guidance and priorities
- With the pandemic “seemingly far from over,” policy support should be sustained and, in some cases, increased.
- Health and containment:
  - Strong health care and containment measures vital until the pandemic has abated.
- Fiscal policy:
  - Targeted fiscal spending needed until recovery is entrenched.
  - Fiscal support should aim at the most vulnerable where fiscal multipliers are highest, and to jobs-oriented, inclusive, and green investment.
  - Credible fiscal plans will be key to secure debt sustainability.
- Monetary and financial stability:
  - Monetary policy should remain supportive.
  - Elevated credit risks demand continual monitoring, especially where debt levels are high.
- Labor market and structural policies:
  - Keep workers connected to the labor force and solvent firms in business while allowing nonviable firms to exit and facilitating new businesses to emerge to mitigate scarring.

### Link to analytical chapters (report structure)
- Chapter 3: containment and related policy measures—fast implementation and appropriately timed exits supported by strong testing and contact tracing key in stabilizing COVID-19 spread while mitigating economic costs; fiscal support critical to reduce economic costs, underpin recovery, and limit scarring.
- Chapter 4: distributional consequences—the crisis hits low-income workers, women, and youth hardest, increasing inequality; medium-term effects could be amplified by automation displacing low-skilled workers; policies should be targeted to mitigate distributional consequences and underpin economic activity and virus containment.

*International Monetary Fund | October 2020*

---

### Growth projections and near-term outlook for Asia and Pacific

- Asia and Pacific region projected to grow by 6.9 percent in 2021.
- Projection for 2021 is 0.3 percentage point higher than projected in June.
- Despite upward revision, projection implies a further drop in the level of output in 2021 than envisaged earlier.
- Asian advanced economies expected to shrink by less than previously projected, reflecting faster pickup after earlier exit from lockdowns.
- Recovery likely sluggish; domestic private sector demand expected to recover slowly due to prolonged social distancing and containment measures.
- Output expected to remain below pre-pandemic trend through the medium term.
- Activity seen as beginning to normalize next year in badly hit emerging market economies.
- Stronger recovery in China, the United States and the euro area will also support growth in Asia.

### Scarring, labor markets, and structural constraints
- Returning to full capacity will be prolonged; fear of infection and social distancing dim consumer confidence and keep activity below capacity until a vaccine is developed.
- International borders likely to remain closed for a considerable period.
- Scarring larger in countries dependent on tourism and in-person services—small states in South Asia, Pacific islands, and some advanced and emerging market economies.
- Labor market signs of scarring:
  - Aggregate hours worked declined as employment rates and hours per employee collapsed.
  - Unemployment surged; labor force participation plunged, particularly for women and younger workers.
- Extent of scarring depends on: reliance on contact-intensive activities; market rigidities; digital penetration, technological capacity, availability of remote working; policy space to support resource reallocation.
- Weak household, corporate, and financial balance sheets may add to scarring.

### Trade prospects and external exposure
- Prospects for a global trade–led recovery are decidedly uncertain.
- China’s recovery can boost regional trade, but weak global growth, closed borders, and tensions around trade, technology, and security worsen prospects for trade-led recovery.
- Some countries started diversifying and moving up the value chain, but reorientation toward domestic demand will take time and policy effort.
- Small economies (Pacific island countries) face exceptional difficulty reorienting away from tourism, commodities, and remittances due to size, remoteness, and natural disaster vulnerability.

### Uncertainties, unknowns, and risks
- Early vaccine success could quicken recovery, but downside risks are considerable.
- Second wave cannot be ruled out.
- Escalating US-China tensions across trade, financial flows, technology, and geopolitics pose major risks.
- Pandemic’s disproportionate impact on the poorest will exacerbate income and wealth inequality in Asia and could engender social tensions.
- Return to tighter financial conditions could exacerbate pre-pandemic vulnerabilities, tip corporations and SMEs into bankruptcy, worsen credit risk and financial stability, and aggravate debt overhangs.

### Policy recommendations: smart, green, inclusive recovery
- Full arsenal of policy support needed; fiscal and monetary support should continue because of evident economic slack.
- Health and containment:
  - Flatten the curve where outbreaks not under control; micro-containment measures, timely testing, effective contact tracing, increased hospital capacity.
  - Credible plans to secure vaccine supplies, including multilateral vaccine sharing efforts.
  - Ramp up health care systems in many emerging market and developing economies, including Pacific island countries, to meet Sustainable Development Goals.
- Fiscal priorities:
  - Spending on health care, targeted social protection, and assistance for viable small and medium enterprises.
  - Better targeting to the most vulnerable, including informal sector, to boost fiscal multipliers (examples cited: Cambodia, India, Indonesia, Vietnam, Nepal).
  - Prioritize green energy and technological infrastructure to create jobs and improve teleworking.
  - Adopt a credible medium-term fiscal strategy, including revenue mobilization and spending efficiency.
  - Low-income countries, including Pacific island countries, will require donor assistance in the form of concessional financing and grants for the foreseeable future.
- Monetary policy:
  - Remain supportive while output gaps large and inflation pressures low: policy rate cuts and unconventional support (Indonesia, Japan, Korea, Thailand); improve policy transmission (strengthen interest rate–based framework in China).
  - In some low-inflation cases, debt monetization could be appropriate if well communicated, limited in size, time-bound, and implemented within a clear operational framework preserving central bank independence.
  - Exchange rates should act as shock absorbers; temporary, well-designed capital flow measures could be considered if flexibility amplifies contraction and heightens financial instability.
- Financial sector:
  - Provide temporary and targeted liquidity support as needed.
  - As recovery takes hold, dial back emergency measures while attending to elevated credit risk and debt overhang.
  - Tighten micro and macroprudential measures as appropriate; ensure financial sector capitalization and provisioning; resolve nonperforming loans promptly.
- Structural reforms:
  - Protect the poor, reduce informality, and reduce worker disengagement and skill erosion.
  - Make safety nets more inclusive of informal workers; facilitate training and redeployment.
  - Incentivize corporate restructuring and consider new equity-like instruments to help viable SMEs overcome debt overhang and retool.
  - Streamline insolvency frameworks to facilitate corporate debt restructuring and resource reallocation.
  - Ease administrative and regulatory barriers for new investment (including foreign investment), exports (especially food and medical supplies), and start-ups.
  - Level the playing field between state-owned and private enterprises to support business formation and job creation (China, India, Indonesia, Vietnam).
  - Improve agricultural productivity and contingency planning; international cooperation and multilateral safety nets (for example, ASEAN+3 Emergency Rice Reserve) important for food security.

*International Monetary Fund | October 2020 — Regional Economic Outlook: Navigating the Pandemic: A Multispeed Recovery in Asia*

---

### Containment, lockdowns, reopening, and empirical evidence

### Containment index and timing
- IMF developed a containment index for six sectors (international travel, schools, retail, industry, services, public gatherings) for 11 Asian countries and 22 European economies; index distinguishes sectors and captures announcements about future changes.
- Asian authorities generally responded early: on average, Asian countries tightened domestic restrictions five days after a significant outbreak (100 cumulative cases); Indonesia waited 25 days.
- Sequencing across countries similar: international travel restrictions first, then school closures.

### Stringency, duration, and reopening experiences
- Stringency and duration differed markedly:
  - Near-complete lockdowns for more than a month: Malaysia, New Zealand, Philippines.
  - Others closed only nonessential services and allowed industry to operate: Australia, Thailand, Vietnam.
  - Korea relied on strong recommendations, voluntary social distancing, and comprehensive testing and tracing rather than mandatory shutdowns.
- Effectiveness of lockdowns varied; implementing/enforcing lockdowns harder in densely populated emerging markets with higher informality and poverty (India, Indonesia, Philippines).
- Reopening:
  - Most reopened after suppressing the virus; most eased restrictions when new cases were more than 80 percent below peak levels.
  - Australia and Japan saw substantial second waves; China, Korea, New Zealand, Vietnam saw small contained outbreaks.
  - India, Indonesia, Philippines reopened before infection rates fell significantly and continued to experience high new infections.
  - Economic activity recovered more slowly in early openers; many sectors remain partially closed.
  - Some countries adopted sequential reopening and reimposed localized lockdowns as needed (China, Vietnam).

### Key lessons from Asia’s experience
- Activate containment early when infection rates low to flatten the curve and reduce depth/duration of economic downturn.
- Exiting lockdowns after virus suppression yields better health and economic outcomes; sequenced approach prioritizing essential sectors and region-based risk assessments reduces economic costs while minimizing health risks.
- Comprehensive testing and tracing systems minimize risk of second waves; adequate testing, effective tracing and isolation, and localized lockdowns reduce community transmission.
- Exact system details vary by country depending on societal preferences and legal privacy protections.

### Impact of containment measures (empirical findings)
- Containment measures effective in flattening the pandemic curve:
  - New Zealand’s very stringent measures likely reduced infections by almost 90 percent relative to baseline of no containment.
  - Containment measures reduced COVID-19 infections by an average of more than 90 percent in 30 days.
- Containment strongly reduced mobility and economic activity:
  - NO2 emissions—proxy for economic activity—cumulatively fell by almost 99 percent 30 days after stringent containment measures, relative to the country-specific path without containment.
  - Translated to economic terms, containment led to about a 12 percent decline (month-on-month) in industrial production.
  - Observed declines in industrial production after lockdowns: China (more than 10 percent) in January–February, Japan (10 percent), Vietnam (15 percent) in April.
- Heterogeneity:
  - Measures more effective where de facto mobility curtailed most (compliance or voluntary distancing).
  - More effective in countries with larger elderly shares and stronger health systems; more challenging in densely populated emerging markets with weaker health systems (example: India).

### Speed of response and modeling results
- Public health response time (days to implement containment after 100 cases) mattered; Asia tended to do relatively well.
- Countries that acted swiftly (Vietnam, Pacific island countries) saw reductions in infections by more than 95 percent relative to no-containment baseline.
- SIR macro model (with fiscal policy) simulations:
  - Delayed containment leads to significantly higher cumulative infections and a more pronounced economic contraction.
  - Modeled delay of containment by 10 weeks produces substantially more weekly infections per million relative to the optimal containment scenario.

### Exit strategies and role of testing/tracing
- Exiting lockdowns yields smaller economic gains than losses from lockdowns:
  - On average, lockdowns led to about a 12 percent monthly contraction in economic activity (industrial production).
  - Full reversal of containment measures would increase activity by only about 6 percent.
- Strong testing and tracing and targeted/localized lockdowns crucial for avoiding infection spikes at exit.
- Easing containment associated with larger infection increases where testing/tracing policies were weaker at exit.
- Asian countries had relatively high testing/tracing indices at the time they eased lockdowns.
- Epidemiological criterion: rollback of strict containment should begin only when new infections are declining, absent herd immunity, reliable vaccines, or effective treatment (WHO 2020).
- Median seven-day average of new cases in Asia at exit was less than 1 per million—among the lowest across regions.
- Countries easing lockdowns when new infections were very low experienced significant mobility and economic activity gains; those reopening when infections were high did not.

*International Monetary Fund | October 2020 — Box 2.1 (continued), Regional Economic Outlook: Navigating the Pandemic: A Multispeed Recovery in Asia*

---

### Industrial production, fiscal mitigation, and multipliers

### Timing of exit and economic effects
- Premature or "early exit" from containment can trigger a second wave and weaken the medium-term recovery.
- Impact measured after 30 days on industrial production (implied by NO2 emissions) to a unitary easing of containment relative to baseline of no change; NO2-to-industrial production elasticity used: 0.015.
- Lighter shade in charts indicates effects not statistically significant at the 95 percent level.
- New cases at time of exit measured per million people, 7-day moving average.
- Regions compared include Asia and Pacific, Africa, Middle East and Central Asia, Americas, and Europe, with Low and High new cases-at-exit scenarios.

### Macroeconomic policy mitigation
- Supportive discretionary fiscal and monetary measures effective in mitigating economic costs of containment.
- Such measures had a much larger impact on economic activity—equivalent to a 22 percent decline in industrial production—in countries with relatively small fiscal packages (text preserved as in source).
- Larger policy rate cuts mitigated some adverse impacts of containment measures.

### Fiscal announcements and estimated multipliers
- Daily database of new announced fiscal plans constructed for 39 advanced and emerging market economies; high-frequency identification used by purging fiscal news with daily NO2 and mobility indicators.
- Estimates suggest:
  - Fiscal announcements of 1 percent of GDP increased year-on-year industrial production by about 0.4 percent—equivalent to a fiscal multiplier of about 0.2–0.3.
  - Fiscal announcements of 1 percent of GDP during months of larger losses in economic activity led to about a 1.2–1.4 percent increase in industrial production (corresponding to a fiscal multiplier of 0.6–1).
- Multipliers higher in economies with fixed exchange rates, in more closed economies, and where debt-to-GDP ratios relatively low.
- Fiscal announcements generally have larger effects when containment measures are more stringent, though controlling for fiscal announcements during months of weaker activity shows bigger impact when containment measures are lower (supply-side restrictions smaller).

### Targeted fiscal transfers and health outcomes
- Model simulations show fiscal measures targeted to the most vulnerable (consumption coupons in Korea; cash transfers to casual workers in Australia) helped:
  - Reinforce greater social distancing.
  - Reduce number of infections and fatalities.
- Targeted fiscal announcements were essential for protecting the vulnerable, stimulating activity, and helping contain the pandemic; they should not be withdrawn prematurely.

### Key data and methodology notes
- Industrial production impacts implied by NO2 changes using elasticity 0.015.
- Panel regression of 39 countries from January 2020 to July 2020 used to estimate fiscal shock impacts.
- Charts compare GDP behavior under early exit versus optimal policy scenario, in percent of GDP.

*Regional Economic Outlook: Navigating the Pandemic: A Multispeed Recovery in Asia, International Monetary Fund | October 2020.*

---

### COVID-19, inequality, automation, and social unrest in Asia

### Labor market impacts and rising inequality
- High-frequency labor surveys indicate inequality increasing because job losses concentrated among low-income workers.
- Aggregate hours worked declined at both extensive (employment rate) and intensive (hours per employee) margins; unemployment surged; labor force participation plunged.
- Job losses concentrated in high-contact sectors (hospitality and retail) and non-teleworkable industries (mining, manufacturing, construction).
- Average monthly wage in the social sector is less than one-third that of essential and teleworkable industries.
- Gender effects:
  - Between December 2019 and June 2020 Asia’s female participation rate declined by 1.3 percentage points compared with a 1 percentage point fall for males.
- Youth effects:
  - Asia’s youth experienced sharper job losses; youth unemployment rose 1.4 percentage points, on average, by June.

### Pandemics, automation, and medium-term inequality
- Historical evidence (Furceri and others 2020): major epidemics over past two decades led to persistent increases in the Gini coefficient, higher income shares for higher-income deciles, and lower employment-to-population ratios for low-education workers.
- Robot adoption (new robot installations per 1,000 employees, International Federation of Robotics) tends to increase after pandemics, especially when associated with significant contraction.
- Increase in inequality over medium term tends to be larger in economies with higher robot density—above 2.3 per thousand.
- In 2018 nearly two-thirds of the world’s operational stocks of industrial robots were in Asia.
- More than 40 percent of the world’s new robots were installed in China.
- Robot density rising fast from a low base in several Asian economies.
- Empirical setup: impulse responses estimated using sample of 14 industries in 39 economies over 2000–14 with local projection method (Jordà 2005). Robust standard errors clustered at country-industry pair level.

### Pandemics, inequality, and social unrest
- Panel VAR evidence (sample of 133 countries over 2001–18) finds:
  - Past major pandemics, by reducing growth and increasing inequality, led to a significant increase in social unrest in the medium term (measured by civil disorder score from International Country Risk Guide).
  - No significant short-term effects found.
- Nonlinearities:
  - Effect of inequality on social unrest stronger when initial income inequality high.
  - Increase in net Gini associated with higher social unrest when net Gini above 40—about one-third of Asian economies have net Gini higher than this threshold.
  - Impact of inequality on unrest depends on extent of redistribution: increase in inequality associated with more unrest when redistributive transfers are low, suggesting redistributive measures reduce social tensions.
- Methodology:
  - Panel VAR uses orthogonalized impulse responses with three endogenous variables: real growth, net Gini coefficient, civil disorder. Pandemic dummy exogenous. Country fixed effects controlled; standard errors clustered at country level. 90 percent confidence bands estimated using Gaussian approximation based on 200 Monte Carlo draws.

### Policy implications and model-based findings
- Panel regression sample: 133 countries over 2001–18; regression form reported in source text.
- Susceptible-infected-recovered macro model (Eichenbaum, Rebelo, Trabandt 2020) extended with skilled/unskilled workers, external borrowing, and redistributive fiscal policy (Engler and others 2020) yields:
  - Fiscal support measures can reduce infections by about one-third relative to no-intervention baseline.
  - Mechanism: fiscal support protects livelihoods, increases disposable income, makes staying home more affordable, reinforcing social distancing.
- Targeted versus untargeted transfers (model comparisons):
  - Targeted transfers (TT) raise GDP by some 3 percent (relative to untargeted transfers).
  - Targeted transfers lower debt-to-GDP ratio by 6 percentage points (relative to untargeted transfers).
  - Targeted support increases consumption share of the unskilled in GDP relative to untargeted transfers and reduces consumption share of the skilled due to redistributive measures.
  - Favorable effects larger for targeted measures because unskilled workers more exposed to health crisis; preserving unskilled consumption saves more lives and flattens debt-to-GDP surge.
- Policy implications:
  - Economically and socially beneficial to provide targeted support to the unskilled.
  - To minimize longer-term damage, policies should address automation challenges:
    - Revamp education curricula for flexible skill sets and lifelong learning.
    - Provide new training for adversely affected workers.
  - No one-size-fits-all policy, but targeted measures dominate untargeted measures in model outcomes.

### Observed policy responses and implementation constraints in Asia
- Advanced economies introduced targeted cash transfers more than emerging market and developing economies.
- Most advanced economies also introduced enhanced unemployment benefits, wage subsidies, and firm-level fiscal support.
- Determinants of implementation:
  - Degree of digitalization aided ability to reach citizens; low-income and emerging market countries that introduced targeted cash transfers (Cambodia, India) had, on average, higher digitalization scores than those that did not.
  - Less frequent adoption of enhanced unemployment benefits, wage subsidies, and firm support among low-income countries and emerging markets likely related to higher informality, making reaching workers and firms more challenging.

*Regional Economic Outlook: Navigating the Pandemic: A Multispeed Recovery in Asia, International Monetary Fund | October 2020.*

### 1.   Overview   1

### 1. Overview

### Global context and headline projections
- The global economy is beginning to recover after a sharp contraction in the second quarter of 2020 as nationwide lockdowns are lifted and replaced with more targeted containment measures.
- Global growth is projected at −4.4 percent in 2020, revised up since the June 2020 World Economic Outlook (WEO) Update because of better-than-expected second quarter outturns in some major countries.
- Global growth in 2021 is projected at 5.2 percent, a little lower than projected earlier, “consistent with the expectation that social distancing persists into 2021 and fades thereafter.”

### Asia and Pacific: multispeed recovery and projections
- The Asia and Pacific region is expected to contract by −2.2 percent in 2020 and to grow by 6.9 percent in 2021.
- These projections are “0.6 percentage point lower and 0.3 percentage point higher, respectively, than in the June 2020 World Economic Outlook Update.”
- The outlook varies by country depending on:
  - infection rates and containment measures;
  - the scale and effectiveness of the policy response;
  - reliance on contact-intensive activities;
  - reliance on external demand.

### Recent developments and country differences
- The pandemic is at various stages across Asia: some countries have contained the first wave; others (Australia, Japan, Myanmar) have experienced second waves; some are still striving to flatten the curve (India, Indonesia, Philippines); most Pacific island countries remain largely free of COVID-19.
- China’s activity trough was in February 2020; recovery was boosted by infrastructure, real estate investment, and a surge in exports (medical/protective equipment and work-from-home electronics), followed by a gradual recovery in private nonhousing investment and consumption.
- In India, activity plunged by 24 percent year-on-year in the second quarter of 2020.
- High-frequency indicators point to a trough in activity for much of Asia in April 2020, with economies recovering thereafter at multiple speeds.
- Advanced economies with lower infection rates showed a bigger pickup in activity through to August 2020 than emerging market and developing economies (excluding China).
- Inflation across Asia has remained largely contained because of a drop in demand, lower oil prices, and stable food prices; inflation has been high in India due to supply-side disruptions related to lockdowns.
- Countries recovering faster typically introduced effective containment measures early and timed their exit well; comprehensive testing and contact tracing infrastructure were key, and fiscal support facilitated the resumption of activity.

### Risks, uncertainty, and scarring
- Forecasts are highly uncertain, with significant downside risks, including:
  - a resurgence of the pandemic;
  - geopolitical tensions—particularly US-China—that may derail the recovery;
  - a rise in social unrest triggered by the pandemic’s disproportionate impact on the poorest and most vulnerable;
  - a return to risk aversion in financial markets that could add to balance sheet vulnerabilities.
- An early, large-scale rollout of an effective vaccine is identified as an upside risk.
- Scarring is likely: labor market participation has fallen, output is expected to remain below pre-pandemic trends over the medium term, and the most vulnerable in society are likely to be hit the hardest.

### Policy guidance and priorities
- With the pandemic “seemingly far from over,” policy support should be sustained and, in some cases, increased.
- Health and containment:
  - Strong health care and containment measures are vital until the pandemic has abated.
- Fiscal policy:
  - Targeted fiscal spending is needed until the recovery is entrenched.
  - Fiscal support should aim at the most vulnerable where fiscal multipliers are highest, and to jobs-oriented, inclusive, and green investment.
  - Looking ahead, credible fiscal plans will be key to secure debt sustainability.
- Monetary and financial stability:
  - Monetary policy should remain supportive.
  - Elevated credit risks demand continual monitoring, especially where debt levels are high.
- Labor market and structural policies:
  - Policymakers need to redouble efforts to keep workers connected to the labor force and solvent firms in business while allowing nonviable firms to exit and facilitating new businesses to emerge and generate new job opportunities to mitigate scarring.

### Link to analytical chapters (report structure)
- Chapter 3 examines containment and related policy measures: fast implementation and appropriately timed exits—supported by strong testing and contact tracing—have been key in stabilizing COVID-19’s spread while mitigating economic costs; fiscal support has been critical to reduce economic costs, underpin recovery, and limit scarring.
- Chapter 4 highlights distributional consequences: the crisis is having the largest impact on low-income workers, women, and youth, increasing inequality; medium-term effects could be amplified by automation displacing low-skilled workers, with potential consequences for social cohesion. Policies should be targeted to mitigate adverse distributional consequences and underpin overall economic activity and virus containment.

*International Monetary Fund | October 2020*

### 8.2 percent in 2021 on the assumption of a

### 8.2 percent in 2021 on the assumption of a

### Growth projections and near-term outlook
- Asia and Pacific region is projected to grow by 6.9 percent in 2021.
- Projection for 2021 is 0.3 percentage point higher than projected in June.
- Despite the upward revision, the projection implies a further drop in the level of output in 2021 than envisaged earlier.
- Asian advanced economies are expected to shrink by less than previously projected, reflecting a faster pickup in activity following earlier exit from lockdowns.
- Recovery is likely to be sluggish; domestic private sector demand is expected to recover slowly due to a longer period of social distancing and containment measures.
- Output is expected to remain below pre-pandemic trend through the medium term.
- Activity is seen as beginning to normalize next year in badly hit emerging market economies.
- The stronger recovery in China, the United States and the euro area will also support growth in Asia.

### Scarring, labor markets, and structural constraints
- Returning to full capacity will be a long slog; fear of infection and social distancing measures are dimming consumer confidence and will keep economic activity below capacity until a vaccine is developed.
- International borders are likely to remain closed for a considerable period.
- Scarring effects are likely larger in countries highly dependent on tourism and other services that require in-person contact, affecting small states in South Asia, the Pacific islands, and some advanced economies and emerging market and developing economies.
- Labor markets show increasing signs of scarring:
  - Aggregate hours worked have declined as both employment rates and hours worked per employee have collapsed.
  - Unemployment has surged, and labor force participation has plunged, particularly for women and younger workers.
- The extent of scarring will depend on: reliance on contact-intensive activities; market rigidities; digital penetration, technological capacity, and availability of remote working; and policy space to support resource reallocation.
- Weak household, corporate, and financial balance sheets may add to scarring.

### Trade prospects and external exposure
- Prospects for a global trade–led recovery are decidedly uncertain.
- Although China’s recovery can boost regional trade, weak global growth, closed borders, and festering tensions around trade, technology, and security have worsened the prospects for a trade-led recovery in the region.
- Some countries have started diversifying their economies and moving up the value chain, but a fundamental reorientation of growth models toward domestic demand will take time and considerable policy effort.
- For small economies (such as the Pacific island countries), size, remoteness, and high vulnerability to natural disasters make it exceptionally difficult to reorient away from tourism, commodities, and remittances.

### Uncertainties, unknowns, and risks
- Early success in developing an effective vaccine could lead to a quicker and stronger recovery, but downside risks are considerable.
- A second wave of the pandemic cannot be ruled out.
- Escalating US-China tensions spanning trade, financial flows, technology, and geopolitics could pose major economic risks.
- The pandemic’s disproportionate impact on the poorest and most vulnerable will exacerbate already high and rising income and wealth inequality in Asia and could engender social tensions.
- A return to tighter financial conditions could exacerbate pre-pandemic vulnerabilities, tip struggling corporations and small and medium enterprises into bankruptcy, worsen credit risk and financial stability, and aggravate debt overhangs.

### Policy recommendations: smart, green, inclusive recovery
- Full arsenal of policy support is needed; fiscal and monetary support should continue because of evident economic slack across the region.
- Strong health care and containment measures remain vital:
  - Redouble efforts to flatten the pandemic curve where outbreaks are not under control.
  - Micro-containment measures, timely testing, effective contact tracing, and increased hospital capacity are necessary.
  - Credible plans to secure adequate vaccine supplies, including multilateral vaccine sharing efforts, are essential.
  - Ramping up health care systems is critical for many emerging market and developing economies, including the Pacific island countries, to meet their Sustainable Development Goals.
- Maintain appropriate fiscal support with an eye toward sustainability and longer-term objectives:
  - Priorities: spending on health care, targeted social protection, and assistance for viable small and medium enterprises.
  - Better targeting to the most vulnerable, including in the informal sector, to boost fiscal multipliers (examples cited: Cambodia, India, Indonesia, Vietnam, Nepal).
  - Prioritize investments in green energy and technological infrastructure to create jobs and improve teleworking opportunities.
  - Adopt a credible medium-term fiscal strategy, including steps to improve revenue mobilization and spending efficiency.
  - Low-income countries, including the Pacific island countries, will require donor assistance in the form of concessional financing and grants for the foreseeable future.
- Monetary policy should remain supportive while output gaps are large and inflation pressures low:
  - Instruments include policy rate cuts and unconventional monetary support (Indonesia, Japan, Korea, Thailand) and steps to improve policy transmission (for example, strengthening the interest rate–based policy framework in China).
  - In some cases where inflation remains low, debt monetization could be appropriate if well communicated, limited in size, time-bound, and implemented within a clear operational framework preserving central bank independence.
  - Exchange rates should continue to act as shock absorbers; temporary and well-designed capital flow measures could be considered if flexibility amplifies contraction and heightens financial instability.
- Financial sector policies:
  - Provide temporary and targeted liquidity support as needed.
  - As recovery takes hold, dial back some emergency measures while attending to elevated credit risk and debt overhang.
  - Tighten micro and macroprudential measures as appropriate; ensure financial sector capitalization and provisioning; resolve nonperforming loans promptly.
- Structural reforms to reduce scarring and boost growth potential:
  - Prioritize measures to protect the poor, reduce informality, and reduce worker disengagement and skill erosion.
  - Make safety nets more inclusive of informal workers and facilitate training and redeployment.
  - Incentivize corporate restructuring and consider new equity-like instruments to help viable small and medium enterprises overcome debt overhang and retool.
  - Streamline insolvency frameworks to facilitate corporate debt restructuring and resource reallocation.
  - Ease administrative burdens and regulatory barriers for new investment (including foreign investment), exports (especially of food and medical supplies), and start-ups.
  - Level the playing field between state-owned and private enterprises to support business formation and job creation (China, India, Indonesia, Vietnam).
  - Improve agricultural productivity and contingency planning; international cooperation and multilateral safety nets (for example, ASEAN+3 Emergency Rice Reserve) are important for food security.

### Containment, lockdowns, and reopening experience in Asia
- The IMF developed a new containment index for six economic sectors (international travel, schools, retail, industry, services, and public gatherings) for 11 Asian countries and 22 European economies; the index distinguishes between key economic sectors and captures announcements about future changes to containment measures.
- Asian authorities generally responded early to the epidemic: on average, Asian countries tightened domestic restrictions five days after a significant outbreak (defined as 100 cumulative cases); Indonesia waited 25 days.
- Sequencing of closures was similar across countries: international travel restrictions first, followed by school closures.
- Stringency and duration of lockdowns differed markedly:
  - Several countries imposed near-complete lockdowns for more than a month (Malaysia, New Zealand, Philippines).
  - Others closed only nonessential services and allowed industrial sectors to continue operating (Australia, Thailand, Vietnam).
  - Korea did not implement mandatory shutdowns; it relied on strong recommendations, voluntary social distancing, and comprehensive testing and tracing.
- Effectiveness of lockdowns in reducing infection rates varied; challenges in implementing and enforcing lockdowns were greater in densely populated emerging markets with higher informality and poverty (India, Indonesia, Philippines).
- Reopening experience:
  - Most reopened after suppressing the virus; most eased restrictions when new cases were more than 80 percent below peak levels.
  - Australia and Japan saw a substantial second wave; China, Korea, New Zealand, and Vietnam witnessed small outbreaks largely contained.
  - India, Indonesia, and the Philippines reopened before infection rates fell significantly and have continued to experience high new infections.
  - Economic activity has recovered more slowly in the early openers; many sectors remain partially closed in those countries.
  - Some countries adopted sequential reopening and reimposed localized lockdowns as needed (China, Vietnam).

*International Monetary Fund | October 2020 — Regional Economic Outlook: Navigating the Pandemic: A Multispeed Recovery in Asia*

### Box 2.1 (continued)

### Box 2.1 (continued)

### Key lessons from Asia’s experience
- Containment measures should be activated early, when infection rates are still low, to effectively flatten the virus curve and reduce the depth and duration of the economic downturn (Chapter 3).
- Exiting lockdowns after the virus has been suppressed leads to better health and economic outcomes. A sequenced approach that prioritizes essential sectors and reopens regions based on forward-looking risk assessments can reduce economic costs while minimizing health risks.
- A comprehensive testing and tracing system can minimize the risk of second waves. Adequate testing, effective tracing and isolation (including quarantining of close contacts and localized lockdowns) reduce community transmission. Exact system details will vary across countries depending on societal preferences and legal protections relating to privacy.

### Impact of containment measures (empirical evidence)
- Containment measures were effective in flattening the pandemic curve:
  - New Zealand’s very stringent measures are likely to have reduced the number of infections by almost 90 percent relative to a baseline of no containment measures (Figure 3.1, panel 1).
  - Containment measures reduced COVID-19 infections by an average of more than 90 percent in 30 days.
- Containment measures strongly reduced mobility and economic activity:
  - NO2 emissions—used as a proxy for economic activity—cumulatively fell by almost 99 percent 30 days after implementation of stringent containment measures, relative to the country-specific path without containment (Figure 3.1, panel 4).
  - Translating this into economic terms, containment led to about a 12 percent decline (month-on-month) in industrial production.
  - Observed declines in industrial production after lockdowns in many Asian countries: China (more than 10 percent) in January–February, Japan (10 percent), and Vietnam (15 percent) in April.
- Heterogeneity in effects:
  - Containment measures were more effective where de facto mobility was curtailed the most (either through compliance or voluntary social distancing).
  - Measures were more effective in countries with a large share of elderly in the population and where health systems were stronger; containment may be more challenging in densely populated emerging markets with weaker health systems (example cited: India).

### Speed of response and modeling results
- Public health response time matters:
  - Public health response time measured as days to implement containment after a significant outbreak (set at 100 cases) played a significant role in flattening the curve. Asia tended to do relatively well on this metric (Figure 3.2, panel 1).
  - Countries that acted swiftly (for example, Vietnam or Pacific island countries) saw reductions in infections by more than 95 percent relative to a baseline with no containment measures (Figure 3.2, panel 2).
- SIR macro model (with fiscal policy) simulations:
  - Delayed containment leads to a significantly higher cumulative number of infections and a more pronounced economic contraction (Figure 3.3).
  - A modeled delay of containment by 10 weeks produces substantially more weekly infections per million relative to the optimal containment scenario.

### Exit strategies, asymmetry of effects, and role of testing/tracing
- Exiting lockdowns yields smaller economic gains than the losses from lockdowns:
  - On average, lockdowns led to a contraction in economic activity (measured by industrial production) of about 12 percent a month.
  - An eventual full reversal of containment measures would increase economic activity by only about 6 percent.
- Testing and tracing matter at exit:
  - Strong testing and tracing policies (for example, Korea) and targeted/localized lockdowns were crucial for avoiding spikes in infections when containment was eased.
  - Easing containment measures was associated with a larger increase in COVID-19 infections in countries with weaker testing and tracing policies at the time of exit.
  - Asian countries had relatively high testing and tracing-policy indexes at the time they eased lockdowns.
- Epidemiological criteria for safe rollback:
  - Health considerations suggest rollback of strict containment should begin only when there are clear signs that new infections are declining, absent herd immunity, reliable vaccines, or effective treatment (WHO 2020).
- Outcomes observed:
  - Median seven-day average of new cases in Asia at the time of exit was less than 1 per million people—among the lowest across all regions.
  - Countries that eased lockdowns when new infections were very low experienced significant increases in mobility and economic activity; countries that reopened when new infections were still high did not.

*International Monetary Fund | October 2020 — Box 2.1 (continued), Regional Economic Outlook: Navigating the Pandemic: A Multispeed Recovery in Asia*

### 1. Industrial Production, Deviation from Baseline

### 1. Industrial Production, Deviation from Baseline

### Timing of Exit and Economic Effects
- Premature or "early exit" from containment can trigger a second wave of infections and weaken the medium-term recovery (Figure 3.5, panel 4).
- Bars show the impact after 30 days on industrial production (implied by changes in NO2 emissions) to a unitary easing of containment measures relative to a baseline of no change. Changes in NO2 emissions are translated into industrial production using estimated historical elasticity of 0.015.
- The lighter shade indicates effects not statistically significant at the 95 percent level.
- New cases at time of exit are measured per million people, 7-day moving average.
- Regions compared include Asia and Pacific, Africa, Middle East and Central Asia, Americas, and Europe, with scenarios of Low new cases at exit and High new cases at exit.

### Macroeconomic Policy Mitigation
- Supportive discretionary fiscal and monetary measures were effective in mitigating economic costs associated with containment measures.
- Such measures had a much larger impact on economic activity—equivalent to a 22 percent decline in industrial production—in countries with relatively small fiscal packages (text phrase preserved as in source).
- Larger cuts in policy rates mitigated some adverse impacts of containment measures (Figure 3.6, panel 1).

### Fiscal Announcements and Estimated Multipliers
- A daily database of new announced fiscal plans was constructed for a sample of 39 advanced and emerging market economies.
- Using high-frequency identification (purging fiscal news by daily indicators of economic activity NO2 emissions, mobility), estimates suggest:
  - Fiscal announcements of 1 percent of GDP increased year-on-year industrial production by about 0.4 percent—equivalent to a fiscal multiplier of about 0.2–0.3.
  - Fiscal announcements of 1 percent of GDP during months of larger losses in economic activity led to about a 1.2–1.4 percent increase in industrial production (corresponding to a fiscal multiplier of 0.6–1).
- Multipliers are higher in economies operating under fixed exchange rates, in more closed economies, and where debt-to-GDP ratios are relatively low.
- Generally, fiscal announcements have larger effects when containment measures are more stringent, though when controlling for the effect of fiscal announcements during months of weaker economic activity, there is evidence of a bigger impact of fiscal news when containment measures are lower (supply-side restrictions smaller).

### Targeted Fiscal Transfers and Health Outcomes
- Model simulations show fiscal measures targeted to the most vulnerable households (for example, consumption coupons in Korea and cash transfers to casual workers in Australia) helped:
  - Reinforce greater social distancing.
  - Reduce the number of infections and fatalities (Figure 3.6, panel 3).
- Targeted fiscal announcements were essential for protecting the most vulnerable, stimulating economic activity, and helping contain the spread of the pandemic, and thus should not be withdrawn prematurely.

### Key Data and Methodology Notes
- Industrial production impacts are implied by changes in NO2 emissions translated using elasticity of 0.015.
- Panel regression of 39 countries from January 2020 to July 2020 used to estimate fiscal shock impacts.
- Charts compare behavior of GDP under early exit versus optimal policy scenario, in percent of GDP.

*Regional Economic Outlook: Navigating the Pandemic: A Multispeed Recovery in Asia, International Monetary Fund | October 2020.*

---

### 4. COVID-19 and Inequality in Asia: Risks of Social Unrest?

### Labor Market Impacts and Rising Inequality
- High-frequency labor surveys indicate inequality is increasing because job losses concentrated among low-income workers.
- Aggregate hours worked declined at both extensive (employment rate) and intensive (hours per employee) margins; unemployment surged and labor force participation plunged.
- Job losses concentrated in high-contact sectors and non-teleworkable industries:
  - High-contact sectors include hospitality and retail (social industries).
  - Non-teleworkable industries include mining, manufacturing, and construction.
  - The average monthly wage in the social sector is less than one-third that of essential and teleworkable industries.
- Gender effects:
  - Between December 2019 and June 2020 Asia’s female participation rate declined by 1.3 percentage points compared with a 1 percentage point fall for males.
- Youth effects:
  - Asia’s youth experienced sharper job losses; youth unemployment rose 1.4 percentage points, on average, by June.
- Regional definitions and data notes:
  - "Asia" in specific figures refers to varying country groupings as stated (examples include Australia, Hong Kong SAR, Indonesia, Japan, Korea, Malaysia, New Zealand, Singapore, Taiwan Province of China, Thailand, The Philippines, and Vietnam, or subsets thereof depending on the chart). Data are seasonally adjusted and based on June 2020 data or latest available.

### Pandemics, Automation, and Medium-Term Inequality
- Historical evidence (Furceri and others 2020) shows major epidemics over the past two decades led to persistent increases in the Gini coefficient, higher income shares for higher-income deciles, and lower employment-to-population ratios for low-education workers.
- Robot adoption (new robot installations per 1,000 employees, International Federation of Robotics) tends to increase after pandemic events, especially when associated with significant economic contraction.
- The increase in inequality over the medium term tends to be larger in economies with higher robot density—above 2.3 per thousand.
- Distributional concerns in Asia:
  - In 2018 nearly two-thirds of the world’s operational stocks of industrial robots were in Asia.
  - More than 40 percent of the world’s new robots were installed in China (as noted in source text).
  - Robot density is rising fast from a low base in several Asian economies.
- Empirical setup:
  - Impulse responses estimated using a sample of 14 industries in 39 economies over 2000–14 with local projection method (Jordà 2005). Robust standard error clustered at country-industry pair level.

### Pandemics, Inequality, and Social Unrest
- Panel vector autoregression evidence (sample of 133 countries over 2001–18) finds:
  - Past major pandemics, by reducing growth and increasing inequality, have led to a significant increase in social unrest in the medium term (measured by civil disorder score from International Country Risk Guide).
  - No significant short-term effects were found (as noted in source).
- Nonlinear effects:
  - The effect of inequality on social unrest is stronger when initial income inequality is high.
  - An increase in the net (post tax and transfer) Gini coefficient is associated with higher social unrest when the level of the net Gini is above 40—about one-third of Asian economies have a net Gini coefficient higher than this threshold.
  - The impact of inequality on social unrest depends on the extent of redistribution: an increase in inequality is associated with more unrest when redistributive transfers are low, suggesting redistributive measures help reduce social tensions.
- Methodological notes:
  - Panel VAR uses orthogonalized impulse response functions with three endogenous variables: real growth, net Gini coefficient, and civil disorder. Pandemic dummy is exogenous. Country fixed effects controlled; standard errors clustered at country level. 90 percent confidence bands estimated using Gaussian approximation based on 200 Monte Carlo draws.

### Policy Implications and Way Forward
- Countries with broader social safety nets, greater fiscal space, lower informality, and higher digitalization were better able to protect the vulnerable.
- Countries that entered the crisis with weaker initial conditions faced greater challenges in mitigating distributional effects.
- Policy recommendations implied by analysis:
  - Maintain targeted fiscal measures to protect vulnerable households and support activity, as these also help contain virus spread.
  - Use redistributive measures to reduce social tensions where inequality is high.
  - Monitor automation trends and design labor-market policies and training to mitigate displacement of low-skilled workers.
  - Balance containment with economic support to avoid second waves and further scarring of labor markets.

*Regional Economic Outlook: Navigating the Pandemic: A Multispeed Recovery in Asia, International Monetary Fund | October 2020.*

### 1. Impulse Response of Civil Disorder to Pandemics2. Marginal Effect of Net Gini Coefficient on Civil Disorder

### Impulse Response of Civil Disorder to Pandemics; Marginal Effect of Net Gini Coefficient on Civil Disorder

### Empirical specification and data
- Panel regression sample: 133 countries over 2001–18.
- Regression form reported:
  y_it =  + 1 · ineq_{i,t−1} + 2 · ineq_{i,t−1} + 3 · controls_{i,t−1} + _i + _t + _{i,t}.
- Dependent variable y_it: measure of social unrest (sign flipped so that an increase in the score indicates more disorder or higher social unrest).
- Inequality measured by: net Gini coefficient.
- Margins plot: shows marginal effects of a 1-point (out of 100) increase in net Gini coefficient on civil disorder at different levels of net Gini coefficient.
- Uncertainty: Ninety percent confidence intervals are included with the point estimates.
- Sources: International Country Risk Guide; and IMF staff calculations.

### Findings on civil disorder and inequality
- The analysis quantifies how a 1-point (out of 100) increase in net Gini coefficient affects civil disorder across levels of net Gini coefficient, with 90 percent confidence bands around point estimates.
- The sign convention is such that higher scores correspond to more disorder or higher social unrest.

### Policy analysis: fiscal measures, targeting, and epidemiological outcomes
- Model used: susceptible-infected-recovered macro model (Eichenbaum, Rebelo, and Trabandt 2020) extended to include skilled and unskilled workers, external borrowing, and redistributive fiscal policy (Engler and others 2020).
- Epidemiological impact:
  - Fiscal support measures can reduce the number of infections by about one-third relative to the no-intervention baseline.
- Mechanism:
  - Fiscal support protects livelihoods and increases disposable income, making staying home more affordable and reinforcing social distancing.
- Targeted versus untargeted fiscal support (model comparisons):
  - Targeted transfers (TT) versus untargeted transfers (UT):
    - Targeted transfers raise GDP by some 3 percent (relative to untargeted transfers).
    - Targeted transfers lower the debt-to-GDP ratio by 6 percentage points (relative to untargeted transfers).
    - Targeted support increases the consumption share of the unskilled in GDP relative to untargeted transfers.
    - Targeted support reduces the consumption share of the skilled due to redistributive measures.
  - The favorable effects are larger for targeted than for untargeted measures because unskilled workers tend to be more exposed to the health crisis; preserving unskilled consumption saves more lives and flattens the surge in the debt-to-GDP ratio.
- Policy implications from model:
  - It is economically and socially beneficial to provide targeted support to the unskilled.
  - To minimize longer-term damage, policies should address challenges from automation, including:
    - Revamping education curriculums to achieve more flexible skill sets and lifelong learning.
    - New training for adversely affected workers.
  - There is no one-size-fits-all best policy, but targeted measures dominate untargeted measures in the model’s outcomes.

### Asia: observed policy responses to COVID-19 and implementation constraints
- Observed actions:
  - Advanced economies introduced targeted cash transfers more than emerging market and developing economies did.
  - Most advanced economies also introduced enhanced unemployment benefits, wage subsidies, and fiscal support to firms.
- Determinants of implementation:
  - The degree of digitalization likely aided the ability to reach citizens in need; low-income and emerging market countries that introduced targeted cash transfers (for example, Cambodia and India) had, on average, higher digitalization scores than those that did not introduce these measures.
  - Less frequent adoption of enhanced unemployment benefits, wage subsidies, and firm-level support among low-income countries and emerging markets was likely related to a higher degree of informality, which made reaching workers and firms more challenging.

*Source: IMF staff calculations and text from the IMF Regional Economic Outlook chapter "Navigating the Pandemic: A Multispeed Recovery in Asia" (October 2020).*

---


_Source: https://www.imf.org/-/media/files/publications/reo/apd/2020/oct/english/text.pdf_