Research & Policy Briefs From the World Bank Malaysia Hub No. 35 May 15, 2020 Costs and Trade-Offs in the Fight against the COVID-19 Pandemic: A Developing Country Perspective Norman V. Loayza The world is experiencing the worst pandemic crisis in one hundred years. By mid-April 2020, more than 80 percent of countries around the world had imposed strict containment and mitigation measures to control the spread of the disease. The economic fallout from the pandemic has been immense, with dire consequences for poverty and welfare, particularly in developing countries. This Brief first documents the global economic contraction and its potential impact on developing countries regarding macroeconomic performance, poverty rates, and incomes of the poor and vulnerable. It then argues that the pandemic crisis may hurt low- and middle-income countries disproportionately because most of them lack the resources and capacity to deal with a systemic shock of this nature. Their large informal sectors, limited fiscal space, and poor governance make developing countries particularly vulnerable to the pandemic and the measures to contain it. Next, the Brief reviews recent epidemiological and macroeconomic modelling and evidence on the costs and benefits of different mitigation and suppression strategies. It explores how these cost-benefit considerations vary across countries at different income levels. The Brief argues that, having more limited resources and capabilities but also younger populations, developing countries face different trade-offs in their fight against COVID-19 than advanced countries do. For developing countries, the challenge is preserving lives and avoiding crushed livelihoods. Different trade-offs call for context-specific strategies. For countries with older populations and higher incomes, more radical suppression measures may be optimal; while for poorer, younger countries, more moderate measures may be best. Having different trade-offs, however, provides no grounds for complacency for developing countries. The Brief concludes that the goal of saving lives and livelihoods is possible with economic and public health policies tailored to the reality of developing countries. Since “smart” mitigation strategies (such as shielding the vulnerable and identifying and isolating the infected) pose substantial challenges for implementation, a combination of ingenuity for adaptation, renewed effort by national authorities, and support of the international community is needed. The lockdowns may be easing, but the fight against the pandemic has not been won yet. People and economies will remain vulnerable until a vaccine or treatment are developed. The challenge in the next few months will be to revive the economy while mitigating new waves of infection. The Economic Contraction prior to the pandemic) of about -5 percentage points (pp) in emerging and developing Asia; -8 pp in emerging and developing Europe; -7 pp in The COVID-19 pandemic has thrown the world into its worst economic Latin America and the Caribbean; -6 pp in the Middle East and Central crisis since the Great Depression. The adverse shock first originated in the Asia; and -5 pp in Sub-Saharan Africa (IMF 2020a; see figure 1). powerhouses of the world economy—China, Europe, and the United States. It has quickly propagated throughout the globe. Because of the The International Monetary Fund (IMF), in its April 2020 World pandemic, the world economy in 2020 is projected to grow 6 to 7 Economic Outlook, warns of “severe risks of a worse outcome” (IMF percentage points lower than otherwise, with 90 percent of countries 2020a). The growth projections cited above are predicated on the experiencing negative growth rates (IMF 2020a). The sharp contraction condition that the restrictions imposed to contain the spread of the will affect most aspects of economic activity, including trade and labor. disease are concentrated over the first half of the year (first quarter for Global trade volume is expected to decline between 13 and 32 percent in China and second quarter for the rest of the world). However, if the 2020 (WTO 2020). Global labor, measured by working hours, is projected lockdowns are extended by 50 percent longer, with financial conditions to decrease by 10.5 percent in the second quarter of 2020, a decline becoming tighter and fiscal burdens becoming heavier, world GDP growth equivalent to 305 million full-time workers (UN 2020). in 2020 could drop an additional 3 percentage points. This severe scenario would imply mass unemployment, firm closures, and possibly For developing countries, the global contraction carries a large debt and financial crises. The dislocation implied by such a large adverse external shock. The demand for exports has plummeted, and contraction would extend well beyond 2020, making the recovery in merchandise exports are expected to decline in 2020 by 8 to 36 percent, subsequent years weak and volatile. depending on the region and projection scenario (WTO 2020). Commodity prices are declining to record lows, with oil and metal prices The Impact on Poverty projected to drop by 40 percent and 13 percent, respectively, in 2020 (World Bank 2020a). International tourism (measured by tourist arrivals The pandemic crisis is bound to have an impact on poverty. Conservative and tourism receipts) is expected to decrease by 20 to 30 percent in 2020 estimates suggest that the economic contraction will push 48 million to (UNWTO 2020). Remittances, an increasingly important source of income 135 million of people to poverty worldwide, with the estimates in developing countries, may suffer a decline of about 20 percent in 2020 depending on the poverty line used (48 million new poor, using the (World Bank 2020b). External finance is drying up, with the largest capital $1.90/day poverty line for all countries; and 135 million new poor, using outflow from developing countries ever recorded (more than US$80 $1.90/day for low- income, $3.20/day for lower-middle- income, billion since the start of the crisis) and spreads on sovereign debt and $5.50/day for upper-middle- and high- income countries). This will increasing by hundreds of basis points (IMF 2020a). make 2020 the first year since 1998 that the global rate of poverty will increase (Mahler et al. 2020; World Bank 2020c). No less important, the pandemic entails a large domestic shock in developing countries, with direct costs related to morbidity, health care, If inequality were to rise, the poverty impact of the economic and uncertainty; and indirect costs related to the containment and contraction would be much worse. For instance, if the Gini coefficient mitigation measures imposed to reduce the spread of the disease, such as increases by 2 percent in all countries, the number of poor would rise by reduced labor, production capacity, and productivity. The combined 83 million to 200 million people (with the larger estimate allowing for external and domestic shocks related to the pandemic will produce an different poverty lines across income groups) (see table 1 and figure 2). If unprecedented systemic contraction in GDP growth in 2020 throughout the risks of a worse growth contraction materialize, the numbers of poor the developing world, with estimated reductions (relative to expectations could increase by an additional 70 percent. Affiliation: Development Research Group, World Bank. Acknowledgement: Valuable direct inputs were provided by Daniel Gerszon Mahler and Christoph Lakner (on poverty measures); Patrick Walker (on epidemiological projections); Young Eun Kim (on mortality statistics and economic loss); Roberto Fattal-Jaef, Tatjana Kleineberg, and Rishabh Sinha (on macroeconomic modelling); Roberto Chang, Aart Kraay, Steven Pennings, and Firas Raad (detailed comments); and Nurlina Shaharuddin and Izzati Ab Razak (general research assistance). Criticisms, insights, and suggestions were given by Zack Barnett-Howell, Damien de Walque, Asli Demirgüç-Kunt, Gabriel Demombynes, Francisco Ferreira, Jed Friedman, Alex Galetovic, Carlos Ganoza, Jorge Luis Guzman, Felipe Jaramillo, Stuti Khemani, Bill Maloney, Aaditya Mattoo, Mushfiq Mobarak, Apurva Sanghi, Sergio Schmukler, and David Wilson. Nancy Morrison provided excellent editorial assistance. Objective and disclaimer: Research & Policy Briefs synthesize existing research and data to shed light on a useful and interesting question for policy debate. Research & Policy Briefs carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions are entirely those of the authors. They do not necessarily represent the views of the World Bank Group, its Executive Directors, or the governments they represent. Costs and Trade-Offs in the Fight against the COVID-19 Pandemic: A Developing Country Perspective Figure 1. GDP Growth Will Plummet in Almost Every Country in 2020 Figure 2. Poverty Will Increase Worldwide because of COVID-19 5 0.07 4 $1.90 $3.20 $5.50 Real GDP growth projections for 2020 made in April, percent ETH UGA SEN 3 0.06 Increase in poverty headcount ratio EGY VNM CIV No change in Gini: TKM 2 TZA UZB IND y=-0.0073x + 0.0445 MLI GHA 0.05 t-stat(x) =-3.40 LIDC CHN 1 TJK KEN 2% change in Gini: IDN MDG PHL 0.04 y=-0.0074x + 0.052 0 EMDE t-stat(x) =-2.89 -2 0 2KOR KWT 4 TCD 6 8 -1 GAB PRY 0.03 CMR AGO PAK ARM AZE COG COD COL MYS -2 SAU WORLD 0.02 KAZ URY -3 SGP AFG BOL ZMB HUN SRB ARE NGA 0.01 JOR MAR -4 QAT KGZ BGR GEO HKG TUN CHL PER POL IRQ JPN 0 -5 ADVANCEDRUS DZA TUR ROU ARG PRI IRN BRA (2.0) (2.5) (3.0) (3.5) (4.0) (4.5) (5.0) BLR ZAF USA NOR -6 CHEFIN CAN SVK $100 $316 $1,000 $3,160 $10,000 $31,600 $100,000 MEX DNK CZE GBR CYP ECU SWE THA IRL GDP per capita, 2018 in current US$ (Log 10 scale in parentheses) -7 DEU BEL AUT AUS SDN NZL FRA NLD EST UKR -8 ESP LTU SVN No change in Gini 2 percent change in Gini PRT -9 LVA Linear (No change in Gini) Linear (2 percent change in Gini) ITA HRV -10 GRC Source: Prepared by Loayza and Shaharuddin based on data provided -11 by Christoph Lakner and Daniel Gerszon Mahler from PovcalNet (World Bank 2020d); IMF 2020a; and Lakner et al. 2019. -12 LBN Note: The country income groups follow the World Bank classification based on annual gross national income (GNI) per capita: low-income, less than $1,025; -13 lower-middle-income, $1,025–$3,995; upper-middle-income, $3,996– $12,375; Real GDP growth projections for 2020 made in January, percent and high-income, more than $12,375. The number of new people in poverty per capita is calculated using different poverty lines based on income groups. High Upper-middle Lower-middle Low Low-income and lower-middle-income countries use $1.90/day and $3.20/day, respectively. Upper-middle-income and high-income countries use $5.50/day. Source: Prepared by Loayza and Shaharuddin based on data from IMF 2019a, The vertical lines represent the thresholds. 2020a. Note: The figure compares GDP growth projections for 2020 made before and after the pandemic struck, in January 2020 and April 2020, respectively. Not only poverty but also incomes of the poor and vulnerable would “World” represents the growth rate for the global economy. “Advanced,” “EMDE,” and “LIDC” represents advanced economies, emerging market, and be affected, especially if the crisis implies an increase in inequality. If the low-income developing economies, respectively, based on the IMF groupings. Gini coefficient increases by 2 percent in all countries, the income growth Data labels uses the International Organization for Standardization (ISO) rate of the bottom 40 percent of the population would drop on average by country codes. The country income groups follow the World Bank classification an additional 2.7 percentage points (ranging from about 1 percentage based on annual gross national income (GNI) per capita: low-income, less than $1,025; lower-middle-income, $1,025–$3,995; upper-middle-income, $3,996– point to 5 percentage points, depending on a country’s initial income $12,375; and high-income, more than $12,375. The 45-degree line indicates no distribution). changes in the projections of 2020 real GDP growth made in January 2020 and April 2020. Being above (below) the 45-degree line indicates projections An increase in the Gini coefficient by 2 percent in a given country is improved (worsened) for real GDP growth in 2020. For countries that had no not unusual during times of crisis (Lakner et al. 2019). The Gini coefficient projections made in January 2020, projections from October 2019 are used. increased by 6 percent between 1998 and 2002 in Argentina, for instance, Table 1. Millions of People Will Cross the Line into Poverty because of COVID-19 (Increase in Headcount Poverty) Million Scenarios $1.90/day poverty line $3.20/day poverty line $5.50/day poverty line Percent change in Gini Percent change in Gini Percent change in Gini Income group/region 0% 1% 2% 0% 1% 2% 0% 1% 2% Income group Low-income 12.618 14.423 16.541 9.906 10.262 10.833 5.015 4.717 4.042 Lower-middle-income 29.554 38.695 54.358 77.767 92.545 110.949 67.591 66.381 67.976 Upper-middle-income 5.677 8.106 10.437 14.605 21.737 28.882 44.667 58.154 73.191 High-income 0.246 0.269 0.274 1.238 1.273 1.388 1.821 2.096 2.381 Region World 48.095 61.493 81.610 103.516 125.817 152.053 119.094 131.348 147.590 East Asia & Pacific 4.393 5.754 7.960 14.150 19.548 26.061 33.854 42.849 53.555 Europe & Central Asia 0.857 1.144 1.376 2.464 3.052 3.679 6.453 7.642 9.089 Latin America & Caribbean 2.656 4.163 5.422 6.090 8.881 11.437 12.408 16.617 20.322 Middle East & North Africa 2.720 3.300 3.655 6.484 7.720 9.039 10.906 11.803 13.109 North America 0.075 0.075 0.075 0.996 0.996 0.996 0.996 0.996 0.996 South Asia 15.228 20.578 31.669 54.229 65.311 78.473 42.793 40.146 39.977 Sub-Saharan Africa 22.165 26.479 31.453 19.103 20.310 22.369 11.685 11.296 10.542 Source: Prepared by Loayza and Shaharuddin based on data provided by Christoph Lakner and Daniel Gerszon Mahler from PovcalNet (World Bank 2020d); IMF 2020a; and Lakner et al. 2019. Note: The $1.90/day, $3.20/day, and $5.50/day poverty lines are used for low-, lower-middle-, and upper-middle- and high-income countries, respectively (shaded rows). The size of the growth contractions is equal to the difference in GDP growth projections for 2020 between April 2020 and January 2020. 2 Research & Policy Brief No.35 Figure 3. Labor Informality is Higher in Poorer Countries 100 UGA TCD Informal labor as a share of total employment, 2016, percent MOZ MLI RWA NGA AGO NER BEN KHM LAO CAF SLE NPL SEN CIV 90 BDI HTI GIN GHA IND COD MDG BFA BGD LSO CMR PNG IDN TGO AFG MMRZMB COG LBR TZAZWE BOL 80 MWI PAK KEN HND MAR MRT NIC GTM YEM SDN LKA East Asia & Pacific 70 TJK VNM JAM PER PHL LBN SLVPRYIRQ NAM Europe & Central Asia PSE THA BWA 60 EGY GEO LBY COL ALB IRN MEX TUN Latin America & Caribbean MNG ECU AZE DOM PAN 50 CHN ARG ARMDZA Middle East & North Africa KGZ BRA MYS CHL OMN 40 JOR POL GRC OECD ZAF CRI RUS TUR 30 UKR BIH ROU MKD KOR KWT ESP QAT South Asia MDA SAU ITA JPN 20 SRB URY SGP BLR KAZ ISR GBR USA Sub-Saharan Africa BGR LTU CAN HUN SVK HKG NLD IRL 10 HRV PRT FRA BEL AUT DNK CZE DEU AUS CHE NOR SWE 0 SVN FIN 5 6 7 8 9 10 11 ($150) ($400) ($1,000) ($3,000) ($8,000) ($22,000) ($60,000) GDP 2015, constant 2010 US$ (Logs: upper numeral. Absolute value: lower numeral) Source: Prepared by Loayza and Shaharuddin based on data from ILO 2018; Loayza and Meza-Cuadra 2018. Note: Data labels use the International Organization for Standardization (ISO) country codes. OECD includes high-income countries that have been members of the Organisation for Economic Co-operation and Development for more than 40 years. Developing countries are grouped in geographic region as presented in the legend. and rose 4 percent in the single year between 2000 and 2001. On the on daily work to pay for their basic household necessities: if they cannot other hand, a synchronized increase of 2 percent in all countries has never work even for short periods of time, their family’s subsistence is at risk. been recorded. It is, however, not an improbable scenario in the current Moreover, extensive labor informality implies that relief and recovery crisis because, first, the adverse shock is both global and extremely large, policies aimed at formal labor (such as increasing unemployment and, second, the shock is hurting low-income workers who are most insurance, reducing payroll and income taxes, and extending paid sick affected by strict social distancing measures, particularly informal workers leave) have very limited effects. Another relevant issue for the resilience (see the following section). The International Labour Organization (ILO) of the economy in the face of containment restrictions is the ability to recently warned that “Some 1.6 billion people employed in the informal work from home, which depends on the share of formal employment but economy—or nearly half the global workforce—could see their also on sectoral concentration and digital connectivity. It is estimated that livelihoods destroyed due to the continued decline in working hours in advanced countries, about 35 percent to 45 percent of jobs can brought on by lockdowns to curb the spread of COVID-19” (UN 2020). plausibly be performed at home; in sharp contrast, for developing According to ILO, in the first month of the crisis, the incomes of informal countries, working from home is possible for only 5 percent to 25 percent workers dropped by 60 percent worldwide. of jobs (Dingel and Neiman 2020; Sanchez et al. 2020; see figure 4). An increase in poverty and a decline in the incomes of the poor and Limited fiscal space. Low- and middle-income countries do not have vulnerable have implications beyond monetary well-being. They can lead sufficient “fiscal space,” that is, the ability to deploy public funds and to a deterioration in education, nutrition, health, and living conditions in resources to counter a large negative shock (Kose, Ohnsorge, and general, with long-lasting consequences (Banerjee and Duflo 2011). For Sugawara 2018). Although developing countries do not have larger example, the number of people at risk of starvation could almost double because of the pandemic crisis, increasing by 130 million people worldwide, according to the World Food Programme (WFP 2020). Figure 4. Lower-Income Countries Have Much Lower Shares of Jobs that Can Be Done at Home Limited Ability to Cope LUX The economic crisis brought about by the pandemic may hurt low- and .5 middle-income countries disproportionately because they lack the Share of jobs that can be done at home CHE resources and capacity to deal with large systemic shocks. For a shock of GBR BEL SWE MLT DNKISL NLD USA NOR the same magnitude, low- and middle-income countries suffer more than EST .4 FIN IRL SVN CYP FRA AUT advanced countries do in terms of worsening poverty, inequality, human MDV LVA LTU RUS ITA DEU capital losses, economic disruption, and uncertainty, with scarring PLW MNE HRV POL GRC PRT HUN SYC CZE ESP consequences that could last for years or decades to come (World Bank .3 BGR SVK ARE EGY URY WSM MUS 2013). For the pandemic crisis, three structural characteristics make TONPHL BRA MNG GEO MKD FJI SRB CHL PAN TUR developing countries especially vulnerable to the economic shock large: KIR LAO BIHLKA MEX ROU DOM .2 SWZ GUY ZMBBLZ informal sectors, limited fiscal space, and poor governance (Loayza and NPL KGZ SLV THA TGO NER GHABOLECU Pennings 2020). HND GTM PAK CIV UGA LBR AFGBGD KHM RWA SLE MMR .1 Large informal sectors. Informality is rampant in developing countries MDG MOZ (Loayza 2018; see figure 3). In the typical developing country, the informal sector employs 70 percent of the labor force, and in some low-income 0 countries this rate can be over 90 percent. Informal workers lack benefits 0 15000 30000 45000 60000 75000 90000 105000 120000 such as unemployment insurance, health insurance, and paid leave. They GDP per capita (purchasing power parity) are highly exposed not only to the health impacts of COVID-19 but also to the containment and mitigation measures to reduce the spread of the Source: Figure reproduced from Dingel and Neiman 2020. disease. Most informal workers, especially the self-employed, depend 3 Costs and Trade-Offs in the Fight against the COVID-19 Pandemic: A Developing Country Perspective Figure 5. Fiscal Deficits Are Projected to Increase Sharply in 2020 Figure 6. Government Effectiveness Is Lower in Poorer Countries because of COVID-19 100 FINHKG SGP CHE 10.0 NLDDNK NOR SWE Fiscal deficit projections for 2020 made in April 2020, percent of GDP Percentile rank of government effectiveness, 2018 CAN JPN NZL DEUAUS USA FRA 90 ARE GBR AUT IRL PRT ISR EST KOR BEL SVN COG 80 MYS CHL LTU LVA ESP MUS CZE CYP 5.0 QAT POL SVK QAT GEO URY 70 JAM CHN HRV CRI HUN ITA ZAF THA GRC IND BWA SAU RWA BGR 60 IDN OMN TTO BHR HND NOR JOR ALBNAM MKD SRB TCD PHL 0.0 VNM KAZTUR ARG COD 50 ARM RUS PAN -15.0 -10.0 -5.0 KOR0.0 5.0 10.0 TUN AZEPER COL KWT GHA MAR LKA MEX PRI CYP MNG CUB SEN ECU ROU ETH KHM HUN 40 KEN BOL BLR DOM BEN COL SGP UKR MDG TZA UZB DZA IRN NER GIN PHL THA MDASLV BRA BFA URY -5.0 UZB CZE CHE BEN ZMB KHM CIV PRY HTI NIC MEX MYS LVA NZL IRL BFA BLR MMR NPL VNMMLI DOM PNG CMR ZWE SWE RUS NLD SVK DEU 30 UGAETH KGZ EGY HND BIH IND CIV IDN CHL KAZ GMB PAK ZMB TJKMDA HRV SVN HKG LBN UGA SEN POL FIN AGO DNK PNG LAO SWZ GTM NGA BGDLAO JPN PRT AUT MWI MRT BGD TZA YEM EGY MOZ LTU 20 NER CMR NIC RWA MAR PERGBR EST ITA ECU MOZ LSO GAB KEN TUR ROU BEL GIN MLI NPL UKR FRA GRC NGA AGO TKM PAK KGZ ESP TGOTJK MMR GHA LKA -10.0 AUS 10 SLE MDG ZWE COG GNQ BRA IRN BDI LBR AFG TCD GNB IRQ ISR KWT COD SDN CHN ARE CAN SOM CAF HTI LBY YEM 0 SSD SAU AZE ZAF (2.0) (2.5) (3.0) (3.5) (4.0) (4.5) (5.0) -15.0 DZA USA $100 $316 $1,000 $3,160 $10,000 $31,600 $100,000 OMN SDN GDP per capita, 2018 in current US$ (Log 10 scale in parentheses) -20.0 East Asia & Pacific Europe & Central Asia Latin America & Caribbean Fiscal deficit projections for 2020 made in October 2019, percent of GDP Middle East & North Africa OECD South Asia Sub-Saharan Africa High Upper-middle Lower-middle Low Source: Prepared by Loayza and Shaharuddin based on data from World Bank 2020e World Development Indicators and World Bank 2019 Worldwide Source: Prepared by Loayza and Shaharuddin based on data from IMF Governance Indicators. 2019b, 2020b. Note: Government effectiveness reflects perceptions of the quality of public Note: Data labels use the International Organization for Standardization (ISO) services; the quality of the civil service and the degree of its independence from country codes. The country income groups follow the World Bank classification political pressures; the quality of policy formulation and implementation; and based on annual gross national income (GNI) per capita: low-income, less than the credibility of the government's commitment to such policies. The estimates $1,025; lower-middle-income, $1,025–$3,995; upper-middle-income, $3,996– are in percentile rank terms. Data labels use the International Organization for $12,375; and high-income, more than $12,375. The 45-degree line indicates no Standardization (ISO) country codes. OECD includes countries that have been changes in the projections of fiscal deficit made in October 2019 and April high-income members of the Organisation for Economic Co-operation and 2020. Being above (below) the 45-degree line indicates projections improved Development (OECD) for more than 40 years. Developing countries are grouped (worsened) for fiscal deficit in 2020. in geographic region as presented in the legend. public-debt-to-GDP ratios than developed ones, their debt is more Different Strategies for Different Contexts subject to exchange rate and maturity risks, their credit rating is lower, and their financial markets are shallower. In addition, a small tax base and A welfare evaluation of different strategies to deal with the pandemic less efficient tax administration mean that income support for people crisis should consider both public health and economic considerations jointly. Success in limiting the fatalities and suffering from the disease whose income is disrupted or sharply curtailed because of the pandemic depends on, first, taking into account the economic and institutional and countercyclical fiscal policy are harder to implement in developing challenges faced by different countries and, second, considering the than developed countries. The global, systemic nature of the shock incentives that people have to comply with social distancing measures implies that most countries will be running large deficits. Fiscal revenues (Chang and Velasco 2020a; World Bank 2020f). are likely to be much lower because of the economic contractions, while fiscal expenditures for efforts to mitigate the effects of the crisis are likely Governments have had difficult choices to make on the best approach to be much higher. It is estimated that advanced countries will run an for their countries to contain the spread of the disease. Surrounded by average deficit of about 11 percent of GDP in 2020, while developing uncertainty as to the threat of the virus, some governments chose strict countries will run an average deficit of 9 percent of GDP, double the lockdowns. They may have been a necessary first line of defense in some previously estimated deficit for 2020 (IMF 2020b; see figure 5). In the countries. But, can lockdowns be the foundation of a sustainable strategy uncertain times of COVID-19, as financial markets engage in a “flight to in developing countries? And, with the benefit of more evidence and quality,” many developing countries will find it much more difficult to time, can governments make better choices? cover their fiscal needs, even with some support from the international The Problem with Indiscriminate Lockdowns community (Hausmann 2020). Indiscriminate lockdowns are less effective and more costly in developing Poor governance. The quality of governance determines a country’s than in advanced countries (Barnett-Howell and Mobarak 2020a; Brown, effectiveness to manage shocks and provide assistance (Khemani 2020; Ravallion, and van de Walle 2020; Loayza 2020a; Ravallion 2020). World Bank 2013). Most developing countries suffer from corruption, lack Lockdowns are ineffective in containing the spread of the disease when of transparency and accountability, low bureaucratic competence, and they are imposed in cities with pervasively overcrowded dwellings and burdensome regulatory systems (Kaufmann, Kraay, and Mastruzzi 2011; neighborhoods. There, instead of social distancing, the result from a see figure 6). To make matters worse, some countries suffer from fragility, lockdown is social compression as people are forced into crowded living conflict, and violence (Signé 2020). Faced with the challenge of COVID-19, quarters. Lockdowns are counterproductive when they produce massive developing country governments may find it hard to implement displacement of people, especially from urban to rural areas, spreading, complicated measures to cope with the crisis. To mitigate the impact of rather than containing, the contagion of the virus. And lockdowns are the crisis with economic policies, they may have to rely on straightforward unproductive when compliance is low, leading to short-lived containment relief and recovery policies, such as providing direct cash transfers and gains and an increased probability of second or third waves of infection. ensuring continuity of public goods and services. To contain the spread of Low compliance results not only from weak enforcement capability but the disease with public health measures, developing country also from the dire need that poor people have to work and make an governments may have to adopt strategies that, first, do not require income. strong state enforcement capacity but rely more on community-level In developing countries, lockdowns can be extremely costly in enforcement and household action, and, second, allow for economic and human terms (Basu 2020). They can lead to mass complementary support by nongovernmental organizations, the private unemployment and business closures. They can put the families of poor 4 sector, and international organizations. and informal workers, especially daily laborers, at the risk of starvation, Research & Policy Brief No.35 crime, and disease. Furthermore, the loss of public and private resources Figure 7. Mitigation and Suppression Strategies Have Different Effects due to lockdowns reduces the ability to provide other vital services in on Mortality Rates in Low-, Middle-, and High-Income Countries health care, education, and safety. In India and Pakistan, for example, there is early evidence that the lockdowns have led to fewer prenatal Mortality rate per 1,000 people, population-weighted 10 visits, more unattended home births, fewer child immunization visits, and lower adherence to cancer and tuberculosis treatments (Livemint 2020; 9 Stop TB Partnership 2020). Most developing-country governments do not 8 have the means to prevent these losses and unintended consequences, especially during a deep recession (Didier et al. 2020; Lakner, Özler, and 7 Van der Weide 2020). 6 average 5 Different Vulnerabilities for Different Demographic Profiles 4 An important aspect of the COVID-19 disease is that it affects the young and old differently. According to current epidemiological evidence, there 3 is a marked difference in the age profile of vulnerable populations: the 2 infection fatality rate for people over 60 years of age appears to be 5 to 100 times higher than those below the age of 60, with substantially higher 1 fatality rates in older age groups (Verity et al. 2020). 0 No intervention Mitigation Enhanced Late Early This is relevant for two reasons. First, the population most vulnerable mitigation suppression suppression is the least economically active; and, conversely, those most economically Low Lower-middle active are at considerably lower risk of serious or fatal disease. A blanket constraint of economic activity, therefore, disables those who can work Upper-middle High (and practice responsible social distancing) without increasing Low, adjusted Lower-middle, adjusted significantly their exposure to health risks. Upper-middle, adjusted High, adjusted Second, developing countries, being at an earlier phase of the Source: Prepared by Loayza and Kim based on data provided by Patrick GT demographic transition, have younger populations. The ratio of people Walker (Imperial College London COVID-19 Response Team). over 60 years of age to total population is 1:20 and 1:10 in the typical Note: Adjusted mortality rates take into consideration that critically ill low-income and middle-income country, respectively. In comparison, this patients may not receive proper attention due to constraints in health care ratio is 1:5 in advanced countries. Therefore, the mortality risk from capacity. We adjust for heterogeneity in health care by using the number of hospital beds as proxy. See Kim and Loayza (2020) for details. COVID-19 is significantly lower in low- and middle-income than in advanced countries (Walker et al. 2019). A qualification is that not only age but also comorbidities determine measures decrease with income level. These results are driven by the severity of COVID-19. On the one hand, low- and lower-middle countries’ age composition, with younger populations in lower-income income countries have a lower prevalence of diseases known to countries being less affected by the pandemic and corresponding exacerbate symptoms of COVID-19, such as cardiovascular disease, mitigation and suppression strategies. chronic respiratory disease, and diabetes, even controlling for age Some caveats on these epidemiological projections should be differences (WHO 2019, 2020b). On the other hand, lower-income considered when applied to low- and middle-income countries. First, the countries have a higher prevalence of infectious diseases and, in some original projections do not control for the quality of public health care. To cases, higher HIV infections, which weaken immunity if unattended. partially address this concern, Kim and Loayza 2020 provide a new These two forces might lead to a different risk profile from that observed in China, Europe, and US (Walker et al. 2019), although it is not clear if this computation that attempts to adjust for higher mortality of critically ill will reinforce or not the age-driven mortality differences projected across patients in countries with lower health care capacity (presented in Figure countries. 7 as dashed lines). Second, the projections assume uniform compliance across countries and do not allow for collateral or unintended damage. The Alternatives to Indiscriminate Lockdowns Stricter and more demanding measures will likely elicit lower compliance (except for short periods of time when repression can be imposed) and, So, are there alternatives to strict and indiscriminate suppression moreover, lead to lives lost from starvation, diminished health care, and, measures? How effective are they in reducing mortality risk across in general, worsened extreme poverty. Low compliance and collateral different countries? And, what social and economic losses are involved? damage would diminish the containment effect of radical suppression This Brief considers two approaches to answer these questions, one from strategies, making them less beneficial than other more moderate epidemiology and the other from economics. strategies (reducing further the slope of the lines for developing countries in figure 7). Insights from Epidemiology Barnett-Howell and Mobarak (2020b) go one step beyond One of the world’s top epidemiological teams, the Imperial College epidemiological projections to estimate the economic loss derived from COVID-19 Response Team, has studied the results of alternative estimated fatalities in different countries and for different mitigation and mitigation and suppression strategies on mortality rates in a worldwide group of countries (Walker et al. 2020). Using a variation of the canonical suppression strategies. They highlight four findings. First, the loss from susceptible-infected-recovered (SIR) model, Walker et al. study the effects inaction to address the disease is enormous for all countries. Second, the of five strategies: (1) unmitigated epidemic; (2) social distancing (reducing loss across all scenarios is higher for better-off countries because of a interpersonal contact by 45 percent); (3) enhanced social distancing combination of higher mortality (due to an older population) and higher (reducing, in addition, interpersonal contacts of people aged 70+ by 60 incomes. Third, all countries benefit from mitigation strategies, but the percent); (4) late suppression (reducing interpersonal contacts by 75 marginal gains decrease when moving from mitigation to suppression percent when the number of fatalities exceeds 1.6 per 100,000 people strategies. Fourth, lower-income countries gain less in moving from no per week); and (5) early suppression (triggered at 0.2 fatalities per intervention to mitigation and still less in moving from mitigation to 100,000 people per week). In the model, the infection fatality rate varies suppression than higher-income countries do. across age groups and is constant across countries (Verity et al. 2020). Therefore, differences across countries are given by their demographic The finding that the economic benefits of more strict suppression structures. The findings of Walker et al. for countries at different income measures decline with income level is particularly relevant. levels are presented in figure 7. Barnett-Howell and Mobarak cite three reasons. First, fatality rates in lower-income countries are lower because their populations are younger. The epidemiological projections show that, first, mortality rates are Second, delaying infections is less useful in countries with significantly lower for countries at lower income levels; second, mortality rates for all lower health care capacity. Third, the opportunity cost (in terms of, for country groups decrease as mitigation and suppression measures instance, work and nutrition) of more drastic measures is higher for become more severe; and, third, the gains in lives saved by more severe poorer people. Barnett-Howell and Mobarak conclude, “Simply put, rich 5 Costs and Trade-Offs in the Fight against the COVID-19 Pandemic: A Developing Country Perspective people can more easily meet their basic needs while social distancing, tracing, and isolation of the infected are conducted, in addition to while a poor person may need to prioritize income-generating shielding of the old and vulnerable, social losses can be reduced further. opportunities to put food on their family’s table” (p. 5). (See Kim and For instance, if all symptomatic people are tested and quarantined, Loayza (2020) for a fuller discussion of related methods and results). fatalities will reduce to one-third and economic loss to one-fourth with respect to the benchmark case. If, in addition, one-third of asymptomatic Insights from Economic Modelling infected people are identified and quarantined, social losses will be reduced to one-twentieth the losses under indiscriminate lockdowns. The analytical and policy-oriented response to the pandemic from economics has been robust, with an expanding number of high-quality So, what implications can be drawn for developing countries? The and relevant papers. Box 1 presents a selection of these papers, focusing most prominent macroeconomic models of the pandemic have been on recent macroeconomic contributions. They embed variations of the calibrated to advanced economies in North America or Europe. canonical SIR epidemiological model in a macroeconomic setup in order Recalibration and application to various low- and middle-income to study the interaction between the dynamics of the pandemic and countries is badly needed. Nevertheless, some qualitative implications economic activities such as work and consumption. The models are then can be derived. First, social welfare losses due to the economic calibrated and simulated to jointly derive results on public health contraction are likely to weigh more heavily than those due to fatalities in outcomes (such as infection and fatality rates) and economic outcomes developing than developed countries. This is because developing (such as losses of employment, consumption, and production). Most countries have younger and poorer populations. Second, compliance with economic models consider two basic externalities: an infection strict measures is bound to be significantly lower in developing countries externality (because people do not fully account for the effect that their than in advanced countries. This is because of lower perceived gains from actions may have on the infection of others) and a congestion externality compliance, inadequate government enforcement, lower social trust, and (because public health care facilities may be jammed by people needing weaker economic policies for relief and recovery. The policy implication is services at the same time). These externalities create a clear role for that mitigation and suppression strategies in developing countries should government intervention to improve social welfare. To assess the be less strict and lengthy than in advanced countries. Third, since optimality of these interventions, some models formulate an explicit economic losses in countries that are already low-income matter social welfare function that depends on losses derived from fatalities and significantly for their welfare, the need for finding cost-effective ways for a decline in economic activity. addressing the pandemic is large and possibly higher than in advanced countries. In these papers, economic agents have been modelled with increasing degree of complexity and realism. Early models assume that all A fourth implication has to do with implementation capability. Some economic agents are homogeneous in terms of productivity and, of the “smart” mitigation strategies that could render great benefits (such importantly, vulnerability to the disease. More recent models allow for as shielding the vulnerable and identifying and isolating the infected) economic agents to be heterogeneous regarding sector of economic pose challenges for implementation even in developed countries. Their activity (such as essential and nonessential); productivity (such as implementation is likely to be more difficult in developing countries; yet whether they can work from home); and demographic structure (such as applying “smart” measures is the only sustainable strategy in the absence of a vaccines or treatment. What is needed is a combination of ingenuity young, middle-aged, and old). for adaptation of “smart” strategies, renewed effort by national Public health interventions are also modelled with increasing realism. authorities, and support of the international community. Early models only allow for governments to impose indiscriminate lockdowns, varying only by length of duration and extent of the The Excruciating Trade-Off between Saving Lives and Saving population. More recent models allow for governments to discriminate Livelihoods between sectors (allowing for essential activities to continue while restricting a fraction of nonessential activity); introduce differential What makes managing the COVID-19 crisis so challenging is that if lockdowns or shielding of vulnerable populations (acknowledging the unattended, it could lead to countless numbers of fatalities—yet, if drastic evidence that the old and comorbid are the most affected by the measures to contain the spread of the disease are imposed, it can disease); and conduct testing, tracing, and isolating of cases of infection. produce a deep and long recession, resulting in devastated livelihoods, extreme poverty, starvation, disease, and conflict. For developing These models introduce political economy aspects in various forms. countries, the trade-off is not just between lives and the economy; rather, One is by recognizing the distributional consequences of imposing it is about preventing deaths from COVID-19 versus avoiding crushed blanket quarantines that benefit and hurt different groups of society livelihoods and deaths flowing from those crushed livelihoods (Basu differently: for instance, the old would benefit more from longer, stricter 2020). lockdowns than the young would. Another form of introducing political The trade-off between saving lives and saving livelihoods is economy considerations is by acknowledging that compliance to excruciating but is also real and unavoidable (Ferreira 2020; Loayza mitigation and suppression measures matter for the choice of optimal 2020a; Economist 2020a). The analysis and estimations presented in social distancing: for instance, in low compliance settings, mitigation and previous sections suggest three conclusions. suppression measures should be more moderate. A third form is by recognizing the feedback loops between good economic policy and the 1. Developing countries have limited ability to cope with the pandemic incentives to comply with regulations: for example, when people expect crisis and related trade-offs. In advanced countries, the that the economy will recover and they will have their jobs back, they are lives-versus-livelihoods trade-off can be eased with immense resources. more likely to heed temporary shutdowns. Several developed countries have committed to spend over 10 percent of GDP to alleviate the effect of the pandemic (IMF 2020b). For example, the One of the latest models, by Acemoglu et al. (2020), studies the United States and Denmark will spend, respectively, about US$5,700 and effects of a comprehensive set of policy measures potentially available to US$7,500 per capita to tackle the crisis. Developing countries, on the policy makers. This study provides some results that can help summarize other hand, not only face limited (and shrinking) financial resources but the lessons from economic models of the pandemic. In their benchmark are also burdened by precarious health systems, overcrowded cities, case, where the only available mitigation measure is a uniform lockdown, informal labor markets, poor governance, and, in some cases, fragility and the optimal policy prescribes a strict and extended quarantine, resulting conflict (Loayza and Pennings 2020). In advanced countries, saving lives in a fatality rate of 1.83 percent of the working-age population and an can be afforded even at great expense. In developing countries, a economic loss of 23.4 percent of GDP. If differential lockdowns across single-minded goal of saving lives from the pandemic is unrealistic and groups are possible, the optimal policy prescribes a targeted shielding of can lead to considerable human losses. the old (and vulnerable) until a vaccine becomes available and lighter social distancing for the rest of the population, allowing them to work. 2. Developing countries face different trade-offs than advanced countries. This targeted policy produces an improvement in both health and Given their limited ability to cope with the pandemic, developing economic outcomes, as the fatality rate and the economic loss drop countries suffer more from the contraction in economic activity required 6 almost to half of those in the benchmark case. Furthermore, if testing, by strict suppression measures. These measures have a higher negative Research & Policy Brief No.35 Box 1. A Selection of Recent Macroeconomic Models on the less radical and shorter measures to mitigate contagion. A utilitarian COVID-19 Pandemic government selects an intermediate suppression strategy by shutting down a fraction of nonessential economic activity and redistributes Several recent high-quality studies model the interaction between public income from those who can work to those who do not. Redistribution is health and economic outcomes and policies with increasing costly, however, and the more costly it is, the less intense optimal sophistication and realism. The following summarizes the contributions suppression should be. The model is calibrated to the United States (a from a selection of these studies. country with an older population and less costly redistribution than in developing countries) and suggests closing down 25 percent of Eichenbaum, Rebelo, and Trabandt (2020) extend the canonical nonessential economic activity for about two months and gradually lifting susceptible-infected-recovered (SIR) epidemiological model, where the restriction over coming months reduces mortality rates by half with transition probabilities across health status are exogenous, to account for respect to no mitigation. This suppression strategy is notably less the effect that work and consumption may have on such probabilities. By extensive than current lockdowns (by half). augmenting the SIR model with macroeconomic features, they can estimate fatality rates and income losses jointly and study the effect of Berger, Herkenhoff, and Mongey (2020) study the introduction of public policy health interventions. In the model, all agents are uniform in targeted antigen testing and quarantine of potentially infectious people. They compare it with a benchmark where quarantine is gradually ordered economic productivity and vulnerability to the disease. People fail to fully for a fraction of the population in the absence of testing. The benchmark internalize that their economic activity can increase the contagion rate. In is calibrated to the prevalent lockdown in the United States in March the presence of this externality, government bodies interested in social 2020. A policy of targeted testing and quarantine allows for noninfectious welfare in terms of both health and economic outcomes should intervene people to participate in economic activity. Its gains are substantial, as it to restrict economic activity and limit the extent of infection and allows for a reduction in both mortality rates and economic contraction. mortality. In the absence of knowledge on who the infected are, the best The model estimates that after one year of targeted testing and policy is to restrict economic activity for a period of time, which reduces quarantine, GDP is 10 percent higher and mortality is cut by half than the mortality rate by one third. This comes at a large cost, however, as under the benchmark scenario of indiscriminate shutdowns. (Mortality suppression measures make the economic contraction three times worse would increase to the benchmark in subsequent years if treatments and than it would have been from the health shock alone. Knowledge of who vaccines are not developed). the infected are greatly eases the trade-off. In the limit, when government knows who the sick are and is able to isolate them, the Acemoglu et al. (2020) study both targeted shielding of the elderly and mortality rate is lowered to the minimum and the additional economic testing/tracing/isolating of the infected as alternatives to uniform loss from “smart” containment is negligible. lockdowns. As in previous models, they use an SIR epidemiological model but augment it to include age-specific risks and differential targeted Alvarez, Argente, and Lippi (2020) present a model similar to policies. In the model economy, there are three types of agents: the Eichenbaum, Rebelo, and Trabandt (2020), with an infection externality young, middle-aged, and old (65+). Although all agents can be infected, and a health care congestion externality. The government intervenes to the old have a much higher risk of serious or fatal disease. Acemoglu et al. eliminate these externalities and optimize a social welfare function that then study optimal policies (which minimize social loss from fatalities and depends on health and economic outcomes. Alvarez, Argente, and Lippi lack of work) under various conditions. The model is calibrated to the use a simple suppression measure that mandates a certain fraction of the United States, but its qualitative results have general relevance. In the population to stay at home, allows for imperfect efficiency of lockdowns benchmark case, where the only available mitigation measure is a (in reducing contagion), and considers improvement (or not) in health uniform lockdown, the optimal policy prescribes a strict and extended care capacity. In their benchmark model, optimal policy suggests a quarantine, resulting in a fatality rate of 1.83 percent of the working-age lockdown of about 60 percent of the population during the first month of population and an economic loss of 23.4 percent of GDP. If differential the epidemic, gradually reduced to 20 percent in the following three lockdowns across groups are possible, the optimal policy prescribes a targeted shielding of the old until a vaccine becomes available and lighter months. Under this lockdown policy, the reduction in GDP is in the order social distancing for the rest of the population, allowing them to work. of 8 percent, but the welfare gains derived from reduced mortality with This targeted policy produces an improvement in both health and respect to no policy intervention are twice as high. Optimal policy economic outcomes, as the fatality rate is reduced to 1 percent and the changes with structural conditions, however. If the lockdown is less economic loss to 12.8 percent of GDP. Furthermore, if testing and tracing effective (as would be in developing countries), the optimal severity are conducted, so that isolation of the infected can be added to shielding should be lower and the duration of the lockdown shorter than in the of the old, social losses can be reduced further. For instance, if benchmark. If health care capacity is not expected to improve (and a testing/tracing/isolating is conducted for all symptomatic people, the vaccine is far in the future), the duration of optimal lockdown should be fatality rate will decline to 0.57 percent and the economic loss to 5.9 reduced to virtually zero. The welfare gain of allowing recovered people percent of GDP. If testing is increased so that about one-third of to go back to work is on the order of 2 percent of GDP. asymptomatic infected people are identified and isolated, the mortality rate will decline to 0.12 percent, the economic loss to 1.2 percent of GDP, The paper by Jones, Philippon, and Venkateswaran (2020) is similar to the and the young and middle-aged will not have to undergo any lockdown. previous two papers but introduces home-based work, which is subject to learning-by-doing. The optimal policy in this model is early suppression Finally, Chang and Velasco (2020b) highlight a feature that has not and interventions to facilitate productivity in home-based work. The received enough attention: the interconnection between public health disadvantages of this and previous models are that, first, they do not and economic policies. The success of suppression measures depends on account for the heterogeneity of economic agents (and corresponding the degree of compliance by the population. Compliance, in turn, distributional effects of blanket suppression measures), and, second, the depends on economic incentives: if the population receive transfers to possibility and benefits of targeted antigen testing and quarantine. help them cope during the lockdown and, importantly, expect that policies will support the economy in the recovery phase, then they will be Glover et al. (2020) study the distributional impact of the epidemic and more motivated to comply with containment and mitigation measures. the policies to contain it. Agents vary by age (young and old), economic This feedback loop creates a role for the government to shape sector (essential and nonessential), and health status (susceptible, expectations through clear and sensible actions, commitment, and infected, and recovered). The old do not work and are more vulnerable to leadership. the disease, while the opposite is true for the young. The old prefer strict Note: This box benefitted from valuable inputs by Roberto Fattal-Jaef, Tatjana and extensive suppression measures, and the young would rather have Kleineberg, and Rishabh Sinha. 7 Costs and Trade-Offs in the Fight against the COVID-19 Pandemic: A Developing Country Perspective Figure 8. “Smart” Measures Can Ease the Trade-Off between Lives and Livelihoods a. Different trade-offs for different countries b. Easing the trade-off with “smart” measures Maximal feasible Maximal feasible control control Higher-income countries Economic loss Economic loss No control Optimal uniform No control lockdown Lower- Optimal income targeted Optimal countries No control lockdown targeted + group lockdown distancing + test & trace 0 0 Deaths Deaths Source: Author’s illustrations adapted from Acemoglu et al. 2020. Note: The figures are presented for illustration purposes only. Economic losses are relative to income levels, with larger losses implying higher development setbacks. The downward sloping portion represents the trade-off between economic losses and fatalities; and the upward sloping portion represents the lose-lose situation of more fatalities and higher economic losses. impact on poverty; associated social ills such as malnutrition, disease, and conduct sufficient relief and recovery policies, they would need the conflict; and human capital formation of a large fraction of children with economic support of the international community. International aid is limited or no access to online learning (Giannini 2020; UNESCO 2020). a global public good to face the pandemic because to be defeated, Yet, the benefits of these suppression measures in reducing mortality and COVID-19 has to be defeated everywhere (Lakner, Özler, and Van der the corresponding economic loss are lower in developing countries Weide 2020; Loayza 2020b). because of both their younger populations and their lower ability to treat critically ill patients. Therefore, the lives-versus-livelihoods trade-off is • Sustainable and cost-effective mitigation. Although there is much likely to be different for countries at different levels of development and uncertainty regarding the science around COVID-19, including its demographic profiles (for an illustration, see figure 8a). Different epidemiology, there is relevant evidence that can help guide trade-offs call for context-specific strategies: for countries with older sustainable and cost-effective mitigation measures. They are populations and higher incomes, more radical suppression measures may especially relevant where lockdowns are ineffective and excessively be optimal; while for poorer, younger countries, more moderate costly. They are also important for countries that are considering exit measures may be best. strategies from their lockdowns to prevent a second wave of infection. Some of these measures consist of improving personal and 3. The terms of the trade-offs can be eased by coordinated economic and public hygiene; compulsory wearing of face masks and other public health policies. Having different trade-offs for developing countries protective devices in public places; encouraging working from home provides no grounds for complacency (Demirgüç-Kunt, Lokshin, and Torre when possible; restricting mass gatherings; shielding vulnerable 2020). The goal of preserving lives and livelihoods is possible with a groups like the elderly and people with certain preexisting conditions; combination of pragmatic and effective economic and public health antibody testing of representative samples to obtain a profile of the policies, tailored to the reality of developing countries (Basu 2020; for an extent of infection and recovery in the population; extensive antigen illustration, see figure 8b). The success of public health measures to fight testing, tracing, and isolating cases of infection; and use of the pandemic requires well-designed economic policies. Compliance with information technologies to gather information on social distancing, social distancing is higher when current economic support is available detection of potential contagion, and compliance with mandated and the prospects for economic recovery are better; likewise, public isolation (Basu 2020; De Walque et al. 2020; Loayza 2020a; Maloney health infrastructure can be improved if sufficient revenues are being and Taskin 2020; The Economist 2020b). Here again, the role of generated by a well-performing economy. international coordination and cooperation is essential: first, to share • Public health care capacity. Measures should be urgently undertaken technologies and best practices; and second, to prevent waves of to increase public health care capacity, procuring emergency hospital infection “imported” from other countries with less effective space, breathing ventilators, medical protective equipment, and mitigation measures. testing kits. Given the global rush to acquire medical equipment, developing country governments would need to rely on cost-effective Conclusion solutions in cooperation with their own private sectors and The world is four months into the worst pandemic crisis in one hundred international aid organizations (WHO 2020a; see also box I.B.5 in years. By mid-April 2020, more than 80 percent of countries had imposed World Bank 2020f). strong social distancing measures to control the spread of the disease • Relief and recovery economic policies. Relief measures should be (Hale et al. 2020). The economic fallout has been immense, with dire implemented during the containment period and recovery measures consequences for poverty and welfare, particularly in developing in the aftermath. They should aim to protect the poor and vulnerable countries. As the restrictions are gradually eased, the challenge will be to (for instance, scaling up both targeted and untargeted cash transfers); revive the economy while mitigating new waves of infection. Having more provide temporary support to affected businesses (for example, limited resources and capabilities but also younger populations, granting wage subsidies and tax reductions); and ensure developing countries face different trade-offs in their fight against macroeconomic stability and the continuity of public services. (For a COVID-19. Sooner rather than later, draconian measures may have to give review of economic policies to mitigate the effects of the pandemic, way to more targeted, self-enforcing, and cost-effective practices to see Didier et al. 2020; Loayza and Pennings 2020, and Özler 2020). reduce the rate of infection. Only then can a careful and gradual economic 8 Since most low- and middle-income countries lack the fiscal space to recovery begin. Research & Policy Brief No.35 References Acemoglu, D., V. Chernozhukov, I. Werning, and M. D. Whinston. 2020. “A Multi-Risk SIR Model with Lakner, C., B. Özler, and R. Van der Weide. 2020. “How Would You Distribute COVID-Response Funds to Poor Optimally Targeted Lockdown.” NBER Working Paper 27102, National Bureau of Economic Research, Countries?” World Bank blog, April 13. https://blogs.worldbank.org/developmenttalk/how-would- Cambridge, MA. https://www.nber.org/papers/w27102. you-distribute-covid-response-funds-poor-countries. Alvarez, F. E., D. Argente, and F. Lippi. 2020. “A Simple Planning Problem for Covid-19 Lockdown.” NBER Livemint. 2020. “How Covid-19 Response Disrupted Health Services in Rural India.” April 27. Working Paper 26981, National Bureau of Economic Research, Cambridge, MA. https://www.livemint.com/news/india/how-covid-19-response-disrupted-health-services-in-rural- https://www.nber.org/papers/w26981. india-11587713155817.html Banerjee, A., and E. Duflo. 2011. Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty. Public Affairs. http://web.archive.org/web/20190414231629/www.pooreconomics.com/. Loayza, N. V. 2018. “Informality: Why Is It So Widespread and How Can It Be Reduced?” Research & Policy Brief No. 20, World Bank. http://documents.worldbank.org/curated/en/130391545228882358/ Barnett-Howell, Z., and A. M. Mobarak. 2020a. “Poor Countries Need to Think Twice About Social pdf/Informality-Why-Is-It-So-Widespread-and-How-Can-It-Be-Reduced.pdf Distancing.” April 10. https://foreignpolicy.com/2020/04/10/poor-countries-social-distancing-coronavirus/. __________. 2020a. “Smart Containment and Mitigation Measures to Confront the COVID-19 Pandemic: __________. 2020b. "Should Low-Income Countries Impose the Same Social Distancing Guidelines as Tailoring the Pandemic Response to the Realities of Developing Countries.” World Bank blog, April 7. Europe and North America to Halt the Spread of COVID-19?" Yale School of Management, New Haven, CT. https://blogs.worldbank.org/developmenttalk/smart-containment-and-mitigation-measures- http://yrise.yale.edu/wp-content/uploads/2020/04/covid19_in_low_income_countries.pdf. confront-covid-19-pandemic-tailoring . Basu, K. 2020. "Who’s Afraid of COVID-19?" Project Syndicate. May 6. https://www.project-syndicate.org/commentary/developing-countries-fear-drives-policy-mistakes- __________. 2020b. “Aid Effectiveness during the COVID-19 Pandemic: This Time It Must Be Better.” World by-kaushik-basu. Bank blog, May 6. https://blogs.worldbank.org/developmenttalk/aid-effectiveness-during-covid-19- pandemic-time-it-must-be-better. Berger, D. W., K. F. Herkenhoff, and S. Mongey. “An SEIR Infectious Disease Model with Testing and Conditional Quarantine.” NBER Working Paper 26901, National Bureau of Economic Research, Loayza, N. V., and C. Meza-Cuadra. 2018. “A Toolkit for Informality Scenario Analysis.” World Bank. Cambridge, MA. https://www.nber.org/papers/w26901. http://www.worldbank.org/informality. Brown, C., M. Ravallion, and D. van de Walle. 2020 “The World’s Poor Cannot Protect Themselves from the Loayza, N. V., and S. Pennings. 2020. “Macroeconomic Policy in the Time of COVID-19: A Primer for New Coronavirus” https://drive.google.com/file/d/1BECC9eHrwqOSw8H_zO-UrXvZpCRf-H0n/view. Developing Countries.” Research & Policy Brief No. 28, World Bank. Chang, R., and A. Velasco. 2020a. “Economic Policy Incentives in the Fight against Pandemics.” VOX: CEPR http://documents.worldbank.org/curated/en/951811585836124198/pdf/Macroeconomic-Policy-in- Policy Portal. https://voxeu.org/article/economic-policy-incentives-fight-against-pandemics. the-Time-of-COVID-19-A-Primer-for-Developing-Countries.pdf. __________. 2020b. “Economic Policy Incentives to Preserve Lives and Livelihoods.” NBER Working Paper Mahler, D. G., C. Lakner, R. A. Castaneda Aguilar, and H. Wu. 2020. “The Impact of COVID-19 (Coronavirus) 27020, National Bureau of Economic Research, Cambridge, MA. https://www.nber.org/papers/w27020. on Global Poverty: Why Sub-Saharan Africa Might Be the Region Hardest Hit.” World Bank blog, April 20. https://blogs.worldbank.org/opendata/impact-covid-19-coronavirus-global-poverty-why-sub- De Walque, D., J. Friedman, R. Gatti, and A. Mattoo. 2020. “How Two Tests Can Help Contain COVID-19 and saharan-africa-might-be-region-hardest. Revive the Economy.” Research & Policy Brief No. 29, World Bank. http://documents.worldbank.org/curated/en/766471586360658318/How-Two-Tests-Can-Help- Maloney, W., and T. Taskin. 2020. “Determinants of Social Distancing and Economic Activity during COVID-19: Contain-COVID-19-and-Revive-the-Economy. A Global View.” https://drive.google.com/file/d/1G89Q3DvSlVTvxlRHFJCA3Ilvz1vgZo_U/view. Demirgüç-Kunt A., M. Lokshin and I. Torre. 2020. “The sooner, the better: The early economic impact of Özler, B. 2020. “What Can Low-Income Countries Do to Provide Relief for the Poor and the Vulnerable non-pharmaceutical interventions during the COVID-19 pandemic.” World Bank Policy Research during the COVID-19 Pandemic?” World Bank blog, March 19. Working Paper (forthcoming). https://blogs.worldbank.org/impactevaluations/what-can-low-income-countries-do-provide- Didier T., F. Huneeus, M. Larrain, and S. L. Schmukler. 2020. “Financing Firms in Hibernation during the relief-poor-and-vulnerable-during-covid. COVID-19 Pandemic.” Research & Policy Brief No. 30, World Bank. Ravallion, M., 2020. “Pandemic Policies in Poor Places,” Center for Global Development, April 24. http://documents.worldbank.org/curated/en/228811586799856319/Financing-Firms-in- https://www.cgdev.org/publication/pandemic-policies-poor-places. Hibernation-During-the-COVID-19-Pandemic. Dingel, J. I., and B. Neiman. 2020. “How Many Jobs Can Be Done at Home?” White Paper, Becker Friedman Sanchez, D. G., N. G. Parra, C. Ozden, and B. Rijkers. 2020 “Which Jobs Are Most Vulnerable to COVID-19? Institute for Economics, University of Chicago. What an Analysis of the European Union Reveals” Research & Policy Brief No. 34, World Bank. https://bfi.uchicago.edu/wp-content/uploads/BFI_White-Paper_Dingel_Neiman_3.2020.pdf . http://documents.worldbank.org/curated/en/820351589209840894/Which-Jobs-Are-Most- Vulnerable-to-COVID-19-What-an-Analysis-of-the-European-Union-Reveals. Economist, The. 2020a. “The Coronavirus Could Devastate Poor Countries.” March 26. https://www.economist.com/leaders/2020/03/26/the-coronavirus-could-devastate-poor-countries Signé, L. 2020. “A New Approach Is Needed to Defeat COVID-19 and Fix Fragile States.” Brookings Future Development blog, April 21. https://www.brookings.edu/blog/future-development/2020/04/21/a- __________. 2020b. “App-based Contact Tracing May Help Countries Get Out of Lockdown.” April 16. new-approach-is-needed-to-defeat-covid-19-and-fix-fragile-states/. https://www.economist.com/science-and-technology/2020/04/16/app-based-contact-tracing- may-help-countries-get-out-of-lockdown. Stop TB Partnership. 2020. “The Potential Impact of the Covid-19 Response on Tuberculosis in High-Burden Countries: Modelling Analysis.” Eichenbaum, M. S., S. Rebelo, and M. Trabandt. 2020. “The Macroeconomics of Epidemics.” NBER Working http://stoptb.org/assets/documents/news/Modeling%20Report_1%20May%202020_FINAL.pdf Paper 26882, National Bureau of Economic Research, Cambridge, MA. https://www.nber.org/papers/w26882.pdf. UN (United Nations). 2020. “Nearly Half of Global Workforce at Risk as Job Losses Increase due to COVID-19: Ferreira, F. 2020. “Is There a Trade-off between Lives and Incomes in the Response to Covid-19?” World UN Labour Agency.” UN News, April 28. Bank blog, March 30. https://blogs.worldbank.org/developmenttalk/there-trade-between-lives-and- https://news.un.org/en/story/2020/04/1062792. incomes-response-covid-19. UNESCO (United Nations Educational, Scientific and Cultural Organization). 2020. “Adverse consequences of Giannini, S. 2020. “Prioritize health and well-being now and when schools reopen.” UNESCO News, April 30. school closures.” UNESCO’s COVID-19 Education Response. https://en.unesco.org/news/prioritize-health-and-well-being-now-and-when-schools-reopen. https://en.unesco.org/covid19/educationresponse/consequences. Glover, A., J. Heathcote, D. Krueger, and J. V. Ríos-Rull. 2020. "Health versus Wealth: On the Distributional UNWTO (United Nations World Tourism Organization). 2020. “Tourism and COVID-19.” March 13. Effects of Controlling a Pandemic." Staff Report 600, Federal Reserve Bank of Minneapolis. https://webunwto.s3.eu-west-1.amazonaws.com/s3fs-public/2020-04/COVID19_NewDS_.pdf. https://www.minneapolisfed.org/research/staff-reports/health-versus-wealth-on-the-distributional- effects-of-controlling-a-pandemic . Verity, R. et al. 2020. "Estimates of the Severity of Coronavirus Disease 2019: A Model-based Analysis." The Lancet Infectious Diseases. Hale, T., S. Webster, A. Petherick, T. Phillips, and B. Kira. 2020. “Oxford COVID-19 Government Response https://www.thelancet.com/journals/laninf/article/PIIS1473-3099(20)30243-7/fulltext Tracker.” Blavatnik School of Government. https://www.bsg.ox.ac.uk/research/research-projects/ coronavirus-government-response-tracker. Viscusi, W. K., and C. J. Masterman. 2017. “Income Elasticities and Global Values of a Statistical Life.” Journal of Benefit-Cost Analysis 8 (2): 226–50. Hausmann, R. 2020. “Flattening the COVID-19 Curve in Developing Countries.” Project Syndicate. March 24. https://www.cambridge.org/core/journals/journal-of-benefit-cost-analysis/article/income- https://www.project-syndicate.org/commentary/flattening-covid19-curve-in-developing-countries- elasticities-and-global-values-of-a-statistical-life/5AE299883F668DCC265C41A377E1E063. by-ricardo-hausmann-2020-03. ILO (International Labour Organization). 2018. Women and Men in the Informal Economy: A Statistical Walker, P. G. T., et al. 2020. “The Global Impact of COVID-19 and Strategies for Mitigation and Picture, 3rd edition. Geneva: International Labour Office. Suppression.” Report 12, 26 March. Imperial College London. https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/documents/publication/wcms_626831.pdf. https://www.imperial.ac.uk/mrc-global-infectious-disease-analysis/covid-19/report-12- global-impact-covid-19/. IMF (International Monetary Fund). 2019a. World Economic Outlook, October 2019: Global Manufacturing Downturn, Rising Trade Barriers. World Bank. 2013. World Development Report 2014: Risk and Opportunity—Managing Risk for https://www.imf.org/en/Publications/WEO/Issues/2019/10/01/world-economic-outlook-october-2019 Development. World Bank. https://openknowledge.worldbank.org/handle/10986/16092. __________. 2019b. Fiscal Monitor, October 2019: How to Mitigate Climate Change. __________. 2019. WGI (Worldwide Governance Indicators). https://www.imf.org/en/Publications/FM/Issues/2019/09/12/fiscal-monitor-october-2019. https://info.worldbank.org/governance/wgi/Home/Reports. __________. 2020a. “The Great Lockdown.” Chapter 1 in April 2020 World Economic Outlook. __________. 2020a. “Implications of COVID-19 for Commodities.” April 2020 Commodity Markets Outlook. https://www.imf.org/en/Publications/WEO/Issues/2020/04/14/weo-april-2020. World Bank. https://openknowledge.worldbank.org/bitstream/handle/10986/33624/CMO-April-2020.pdf. __________. 2020b. Fiscal Monitor, April 2020: Policies to Support People during the COVID 19 Pandemic. __________. 2020b. “World Bank Predicts Sharpest Decline of Remittances in Recent History.” April 22. https://www.imf.org/en/Publications/FM/Issues/2019/09/27/Fiscal-Monitor-April-2019-Curbing- https://www.worldbank.org/en/news/press-release/2020/04/22/world-bank-predicts-sharpest- Corruption-46532. decline-of-remittances-in-recent-history. Jones, C. J., T. Philippon, and V. Venkateswaran. 2020. “Optimal Mitigation Policies in a Pandemic: Social Distancing and Working from Home.” NBER Working Paper 26984, National Bureau of Economic __________. 2020c. Macro Poverty Outlook. World Bank. Research, Cambridge, MA. https://www.nber.org/papers/w26984. https://www.worldbank.org/en/publication/macro-poverty-outlook. Kaufmann, D., A. Kraay, and M. Mastruzzi. 2011. “The Worldwide Governance Indicators: Methodology and __________. 2020d. PovcalNet. http://iresearch.worldbank.org/PovcalNet/home.aspx. Analytical Issues.” Hague Journal on the Rule of Law 3 (2): 220‒46. __________. 2020e. WDI (World Development Indicators). https://doi.org/10.1017/S1876404511200046. https://datacatalog.worldbank.org/dataset/world-development- indicators. Khemani, S. 2020. “An Opportunity to Build Legimacy and Trust in Public Institutions in the Time of COVID-19” Research & Policy Brief No. 32, World Bank. __________. 2020f. World Bank East Asia and Pacific Economic Update, April 2020: East Asia and Pacific in http://documents.worldbank.org/curated/en/355311588754029852/An-Opportunity-to-Build- the Time of COVID-19. World Bank. http://hdl.handle.net/10986/33477. Legitimacy-and-Trust-in-Public-Institutions-in-the-Time-of-COVID-19 WFP (World Food Programme). 2020. “WFP Chief Warns of Hunger Pandemic as COVID-19 Spreads Kim, Y. E, and N. V. Loayza. 2020. “Economic Loss from COVID-19 Fatalities Across Countries: A Value of (Statement to UN Security Council).” April 21. Statistical Life Approach” Mimeo, World Bank. https://www.wfp.org/news/wfp-chief-warns-hunger-pandemic-covid-19-spreads-statement-un- https://drive.google.com/drive/folders/1TVOYGZ3Yzvu9zdQIySWNRHDAucjq8P1J security-council. Kose, A., F. L. Ohnsorge, and N. Sugawara. 2018. “Fiscal Space: Concept, Measurement, and Policy WHO (World Health Organization). 2019. “World Health Statistics 2019: Monitoring Health for the SDGs, Implications.” Research & Policy Brief No. 19, World Bank. Sustainable Development Goals.” Geneva. http://documents.worldbank.org/curated/en/771101542638116256/pdf/Fiscal-Space-Concept- https://www.who.int/gho/publications/world_health_statistics/2019/en/. Measurement-and-Policy-Implications.pdf _______. 2020a. “Country & Technical Guidance–Coronavirus Disease (COVID-19).” Lakner, C., D. G. Mahler, M. Negre, and E. B. Prydz. 2019. “How Much Does Reducing Inequality Matter for https://www.who.int/emergencies/diseases/novel-coronavirus-2019/technical-guidance. Global Poverty?” Policy Research Working Paper 8869, World Bank. _______. 2020b. “Coronavirus: Overview.” https://www.who.int/health-topics/coronavirus.WTO (World http://documents.worldbank.org/curated/en/328651559243659214/pdf/How-Much-Does- Trade Organization). 2020. “Trade Set to Plunge as COVID-19 Pandemic Upends Global Economy.” Reducing-Inequality-Matter-for-Global-Poverty.pdf. April 8. https://www.wto.org/english/news_e/pres20_e/pr855_e.htm 9