Research & Policy Briefs From the World Bank Malaysia Hub No. 40 January 7, 2021 The Spread of COVID-19 and Policy Responses Ergys Islamaj, Young Eun Kim, and Duong Trung Le Since early 2020, the COVID-19 pandemic has spread to most countries and territories around the world. For many countries, the second wave of infections is turning out to be more serious than the first. Notwithstanding the global spread of the virus, public policy responses have varied across countries and regions. This brief analyzes the spread of COVID-19 and the effectiveness of policy efforts to contain the disease across a large number of countries. The findings suggest that public health measures—especially testing—and economic support policies are associated with effective containment of the disease, and thus are supporting fundamental prerequisites for a resumption of normalcy, until vaccines are rolled out and community immunity is obtained. This brief examines the evolution of COVID-19 and public policy cases in Sub-Saharan African countries during July, but this has responses across country groups around the world; presents an been largely contained. In many developing countries, the actual econometric analysis of the relationship between the spread of infection rate is likely to be much higher than the reported rate infections and the policy responses; and concludes with main considering weak disease surveillance and reporting capacity (NTI policy implications. and Johns Hopkins University 2019). The case fatality rate—defined as the ratio of the number of The Spread of COVID-19 around the World deaths over cases—peaked in April as the spread of virus stretched COVID-19, which started in China at the end of 2019, spread health capacity in many countries (figure 1, panel b). It has since quickly around the world in 2020 (map 1). The infection converged to below 3 percent. There are multiple potential rate—defined as the number of newly confirmed cases per 1,000 reasons for the decline, including a demographic change in people—has been increasing rapidly in many countries across all infected people toward the young, who are generally healthier regions (figure 1, panel a). Driven by the United States, the and less likely to die; more comprehensive testing; and improved infection rate in North America started to increase rapidly in April. treatment strategies for COVID-19 (Lancet Infectious Diseases Following a transient cool-down period from August to September, 2020). Average case fatality rate remains high in Latin America and new infections have surged again since then. In many European the Caribbean, the Middle East and North Africa, and Sub-Saharan countries, the second wave of infections started in September and Africa, likely due to weaker health care capacity in the countries in is ongoing, with higher infection rates than the first wave in April. these regions (figure 2). In Latin America and Caribbean, the infection rate rose noticeably during winter in the Southern Hemisphere between June and Unabated Spread of Infection August before subsiding, and has declined gradually since. In To better highlight the evolution of new infections across South Asia, the infection rate, after increasing for several countries, this brief classifies countries into three groups: consecutive months, has been declining since October. The • Least affected: The daily infection rate has been consistently infection rate in Middle East and North Africa has been steadily below a constructed threshold. increasing, although it is smaller. Many East Asia and Pacific • Under control: The daily infection rate had been higher than the countries have been less affected relative to countries in other threshold but has been successfully contained below the regions. However, the number of new cases is still high in threshold as of November 2020. Indonesia and the Philippines and has been increasing since • Ongoing: The daily infection rate has yet to be reduced below September in Malaysia and Myanmar. There was a mild surge of the threshold. Map 1. Monthly Infection Rates of COVID-19 by Country COVID-19 has been spreading worldwide since early 2020. a. February b. June c. October Source: Authors’ calculations based on daily data from the European Centre for Disease Prevention and Control (CDC). Note: The maps show the distribution by quintiles in the monthly number of cases per 1,000 people across the three months (low to high = light to dark blue). Affiliations: Ergys Islamaj, East Asia and Pacific Chief Economist Office, World Bank; Young Eun Kim, East Asia and Pacific Chief Economist Research Center, World Bank; Duong Trung Le, East Asia and Pacific Chief Economist Office, World Bank. Acknowledgements: The authors thank Vera Kehayova, Norman V. Loayza, Aaditya Mattoo, Tobias Pfutze, Nurlina Shaharuddin, and Aparnaa Somanathan for useful comments and suggestions. Nancy Morrison provided 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. The spread of COVID-19 and policy responses Figure 1. Rates of COVID-19 Infections and Fatalities by Region The spread of COVID-19 continues unabated in many regions. a. Biweekly infection rate b. Biweekly case fatality rate .02 .04 .06 .08 .1 .12 .14 .16 6 Number of cases per thousand 5 Deaths per case 4 3 2 1 0 0 14 28 11 25 -10 -24 07 21 05 19 02 16 30 14 28 11 25 08 22 06 20 03 17 14 28 11 25 -10 -24 07 21 05 19 02 16 30 14 28 11 25 08 22 06 20 03 17 n- an- eb- eb- ar ar Apr- Apr- ay- ay- un- un- un- uly- Jul- ug- ug- ep- ep- Oct- Oct- ov- ov- n- an- eb- eb- ar ar Apr- Apr- ay- ay- un- un- un- uly- Jul- ug- ug- ep- ep- Oct- Oct- ov- ov- Ja J F F M M M M J J J J A A S S N N Ja J F F M M M M J J J J A A S S N N EAP ECA LAC EAP ECA LAC MENA NA SA MENA NA SA SSA SSA Source: Authors’ calculations based on daily data (January 1–November 17, 2020) from European CDC. Note: Panel a shows total cases per 1000 people for a two-week period for each region. Panel b shows total deaths over total cases for a two-week period for each region. EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin American and Caribbean; MENA = Middle East and North Africa; NA = North America; SA = South Asia; SSA = Sub-Saharan Africa. The threshold is constructed as the 10th percentile of (figure 3). The rise in infections in this group of countries continues country-specific highest daily infection rate for the period January 1 to drive the global rise in COVID-19 cases. In countries in the to November 17, 2020 (0.003 cases per thousand). To smooth daily under-control group, the median infection rate increased from May infection rates, the 15-day moving average is used. A break-down of to July to around 0.06 per thousand but has since declined. In the countries in each group is presented in the appendix. countries in the least-affected group, the median infection rate has remained below 0.005 per thousand since early 2020. The median infection rate in countries in the ongoing group has been rising steadily since March, reaching 1.4 cases per thousand The countries in the least-affected group make up less than 10 population biweekly in November, with no signs of deceleration percent of all countries with available data and are predominantly Figure 2. Health Care Capacity Compared to COVID-19 Fatalities by Figure 3. Biweekly Number of New Cases per Thousand People Region A region with weaker health care capacity tends to have a The infection rate continues to rise in a large group of countries higher case fatality rate, on average. that have yet to control the spread of the virus. 0.10 .07 1.4 0.09 Infection rate (ongoing group) .06 1.2 0.08 .05 1 Infection rate 0.07 Case fatality rate 0.06 .04 .8 0.05 .03 .6 0.04 .02 .4 0.03 .01 .2 0.02 0 0 0.01 14 28 11 25 -10 -24 07 21 05 19 02 16 30 14 28 11 25 08 22 06 20 03 17 n- n- b- b- r r r- r- y- y- n- n- n- y- l- g- g- p- p- t- t- v- v- 0.00 Ja Ja Fe Fe Ma Ma Ap Ap Ma Ma Ju Ju Ju Jul Ju Au Au Se Se Oc Oc No No 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 Global Health Security Index, 2019 Least affected (left scale) EAP ECA LAC MENA NA SA SSA Under control (left scale) Ongoing (right scale) Source: Authors’ calculations based on the country-specific overall Global Health Security Index from NTI and Johns Hopkins University 2019 and data on cases and deaths from European CDC. Note: The figure shows biweekly case fatality rate during November 4–17, 2020 Source: Authors’ calculations based on daily data (January 1–November 17, versus the 2019 Global Health Security Index score for each country with 2020) from European CDC. available data (0 to 100 = weakest to strongest). EAP = East Asia and Pacific; ECA Note: The figure shows biweekly infection rates as a median of country-specific = Europe and Central Asia; LAC = Latin American and Caribbean; MENA = Middle infection rates for each group. Infection rate is defined as the total number of East and North Africa; NA = North America; SA = South Asia; SSA = Sub-Saharan new cases per 1,000 people. See the appendix for a break-down of countries by Africa. group. 2 Research & Policy Brief No.40 Table 1. Status of Infection Rates as of Mid-November 2020 Widespread infection is ongoing in almost three-quarters of all countries. Under control Ongoing Not or Region N mildly Wave(s) suffered Wave(s) suffered affected Total Total Decelerating 1 wave 2+ waves 1 wave 2+ waves trend a East Asia and Pacific 20 7 5 4 1 8 5 3 1 Europe and Central Asia 49 1 0 0 0 48 14 34 17 Latin America and Caribbean 31 0 3 2 1 28 21 7 19 Middle East and North Africa 20 1 2 2 0 17 11 6 9 North America 3 0 0 0 0 3 0 3 1 South Asia 7 0 1 1 0 6 3 3 2 Sub-Saharan Africa 44 8 25 23 2 11 7 4 7 Total 174 17 36 32 4 121 61 60 56 Source: Authors’ calculations based on daily data (January 1–November 17, 2020) from European CDC. Note: The number of waves were determined by the authors in reference to the country-specific infection-rate figures (See the appendix for the URL to access the figures). a. “Decelerating trend” refers to countries where the daily infection rate has been decreasing recently. in East Asia (such as Cambodia, Lao PDR, and Vietnam), Pacific, and Enacting Mobility Restrictions Sub-Saharan Africa (such as Chad, Tanzania, and Togo) (table 1). The countries in the under-control group comprise around 20 To control COVID-19, countries have imposed a variety of mobility percent of countries, more than half of which are in Sub-Saharan restrictions, such as closing schools and workplaces, limiting Africa. The remaining majority, more than 70 percent of the domestic and international travel, and restricting public gatherings countries, are part of the ongoing group, where the number of and events. The stringency of these measures reached the peak in new infections continues to rise or remains high. A significant April as the virus spread in most countries and gradually eased proportion of countries in East Asia and Pacific (60 percent) and in afterward. Existing literature shows a wide variation in the Sub-Saharan African (75 percent) have either been minimally effectiveness of lockdown measures across countries and regions. affected or have been able to control the transmission of For instance, Askitas, Tatsiramos, and Verheydan (2020), Bonardi infections. It is notable that the lower capacity to test and trace probable cases in low-income countries, such as those in Figure 4. Number of Tests per Confirmed Case Sub-Saharan Africa, may have resulted in an underestimate of actual infections. Also, previous experiences with other Countries that have successfully controlled the spread of coronaviruses (such as SARS and MERS) may have made some of infection have a high number of tests per case. these countries better prepared to deal with the current 350 pandemic. 300 Within the ongoing group, the trend of infections was Tests per confirmed case decelerating in almost half of the countries as of mid-November 250 2020 (table 1). Brazil and India were experiencing a decline in the infection rate, while the rates were rising in the United States, 200 Russia, and Mexico. By mid-November, two-thirds of Europe and 150 Central Asia countries were struggling with an accelerating spread of the disease. Around 60 percent of Latin America and Caribbean 100 countries (including Bolivia, Chile, and Peru) were experiencing a decreasing infection rate by November, after a sharp increase 50 during the winter months in the Southern Hemisphere. 0 Most countries in Europe, Central Asia, and North America Least affected a Under control b Ongoing c have encountered a second wave of infections (table 1). Only seven countries have gone through more than one wave and Source: Authors’ calculations based on daily data for infection cases from successfully contained it: Australia (East Asia and Pacific); European CDC and tests from Hasell et al. (2020) for the period January 1–November 17, 2020. Barbados (Latin America and Caribbean); Djibouti (Middle East Note: Bars show the total number of tests per confirmed case (total tests divided and North Africa); Bhutan (South Asia); and Congo, Comoros, and by total confirmed infection cases) by containment group over the period January 1–November 17, 2020. Comparison of the number of tests is affected by the way Guinea (Sub-Saharan Africa). countries report data (https://ourworldindata.org/coronavirus-testing#the-positive -rate-a-crucial-metric-for-understanding-the-pandemic). EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin American and Caribbean; Public Health and Economic Support Policies MENA = Middle East and North Africa; NA = North America; SA = South Asia; SSA = Sub-Saharan Africa. Public policy responses to contain the spread of the disease have a. For the least-affected group: EAP, 405 (average); SSA, 5; other regions, data not varied across countries and have included imposing mobility available. b. For the under-control group: EAP, 975 (average); SSA,10; other regions, data restrictions; testing for infections; and increasing spending to not available. strengthen public health systems, support households, and help c. For the ongoing group: EAP, 15 (average); ECA, 17; LAC, 2; MENA, 4; NA,16; SA,14; SSA,8. firms avoid bankruptcy. 3 The spread of COVID-19 and policy responses et al. (2020), and Weber (2020) argue that the closure of borders is much lower, at 15, in the ongoing group. In other regions, the or travel restrictions has had little effect. In contrast, studies on testing coverage is below 20 on average across the least-affected, international air travel (Chinazzi et al. 2020; Keita 2020) find under-control, and ongoing groups. sizeable effects, particularly if measures were implemented early. Eckardt, Kappner, and Wolf (2020) find limited effectiveness of The importance of intensive testing accompanied by rigorous border controls during the first wave of COVID-19 in 18 Western contact tracing has been emphasized since the early stages of the European countries. pandemic (Rae and Friedman 2020; WEF 2020). Ranan-Eliya et al. (2020) use a sample of 174 countries and show that intensive Challenges in implementing lockdown measures effectively, testing has the greatest impact on controlling the spread of especially in economies facing capacity constraints, are likely to be COVID-19 compared to various other interventions, including mask compounded by unfavorable socioeconomic factors. For instance, usage, school closures, and restrictions on gatherings, and is the many developing countries have a large share of firms in the common characteristic among countries that successfully have informal sector, which usually lacks comprehensive social controlled the disease. Many developing countries have a weak protection. Many governments also face limited fiscal space to capacity for early detection and testing of probable infection cases, provide financial support to low-income earners to sustain as suggested by a low testing coverage or missing data (figure 4). stay-at-home restrictions (Loayza et al. 2020). Lockdown measures such as closing schools can also lead to the loss of human capacity Economic Support in the long term. Many countries have supported households through various Testing measures such as salary subsidies, relief from contractual obligations and debt, and conditional cash transfers. Such Governments across all regions have ramped up their testing effort economic support is likely to induce people to stay at home and since the outbreak of COVID-19. In practice, intensive testing avoid having to look for jobs, which is especially important in policies do not necessarily mean that implementation is rigorous developing countries, where many people working in the informal and effective. The number of tests per confirmed case (testing sector and their families depend on daily income to make ends coverage afterward) has varied widely across regions. On average, meet (Loayza and Meza-Cuadra 2018). Based on the authors’ the countries in the ongoing group have the lowest coverage, at estimations using data from Oxford COVID-19 Government around 12 tests per case; followed by 79 tests per case for the Response Tracker 2020 (Hale et al. 2020), countries in North countries in the least-affected group. The highest coverage, at America and Europe have offered the highest average level of around 300 tests per confirmed case, on average, is reported by income support by subsidizing more than 50 percent of the lost countries in the under-control group that have successfully dealt salary for people employed in the formal sector since April. with the pandemic (figure 4). This trend is driven largely by East Countries in East Asia and Pacific and Latin America and Caribbean Asia and Pacific. Many countries in this region document the have introduced policies to subsidize earnings lost for both highest tests per case globally, including Thailand (313), Australia informal and formal employment on average since May, reflecting (340), Vietnam (983), and China (1,746). The average testing a high share of informal employment, exceeding a median of 60 coverage in East Asia and Pacific region is 405 in the least-affected percent at the regional level (World Bank 2020). Governments in group and 975 tests per case in the under-control group, whereas it countries of the ongoing group have subsidized, on average, less Figure 5. How Public Health and Economic Support Affects COVID-19 Infection Rates Open testing, economic support, and lockdown policies are associated with a lower infection growth rate. a. Open testing policy a b. Economic support b c. Lockdown measures c 0.1 0.1 0.1 cumulative infection growth rate cumulative infection growth rate cumulative infection growth rate 0 0 0 -0.1 -0.1 -0.2 Change in Change in Change in -0.2 -0.1 -0.3 -0.3 -0.4 -0.2 -0.4 -0.5 -0.5 -0.6 -0.3 -0.6 -0.7 1 10 20 30 40 50 60 1 10 20 30 40 50 60 1 10 20 30 40 50 60 Days since open testing policy is Days since a maximum increase in Days since a maximum increase in implemented Economic Support Index Lockdown Stringency Index Source: Authors’ estimations using data on policy responses from Oxford COVID-19 Government Response Tracker (Hale et al. 2020) and data on daily confirmed cases from Europe CDC for January 1–October 31, 2020. Note: This figure presents the point-estimates from regressing cumulative growth rate in new cases on open testing policy (proxied indicator for public-health measure) (panel a) and availability of economic support (panel b). The model controls additionally for country-specific fixed effects, day-specific fixed effects, and the level of countries’ mobility restriction (proxied by lockdown stringency index) (panel c). The Huber-White robust standard error estimation is used. Whiskers represent 95-percent confidence intervals of the estimates. The dependent variable is cumulative infection growth rate, standardized as a unit of deviation from the global mean. a. The testing policy index is constructed as a binary indicator that equals 1 if testing is open and available to all, and 0 otherwise, based on data on testing policy from Oxford COVID-19 Government Response Tracker 2020. b. The economic support index is a continuous average of two indicators: income support for lost earnings and debt relief or deferral on financial obligations, rescaled from [0-100] to [0-1] for representation on the figure. The index is from Oxford COVID-19 Government Response Tracker 2020. c. The lockdown stringency index is constructed as an average of seven indicators with adjustment for whether they are targeted or general: closing of schools, workplaces, and public transport, restrictions on public events, gathering sizes, and domestic and international travel. The methods in Hale et al. 2020 were applied. The index ranges from 0 to 100 corresponding to the least to most intense. 4 Research & Policy Brief No.40 than a half of lost earnings for both formal and informal Countries with infection under control (36) employment since April. In the least-affected and East Asia and Pacific (EAP): Australia, Brunei, China, New Zealand, under-controlgroups, the average income support has been Singapore extended largely to both formal and informal employment over Latin America and the Caribbean (LAC): Barbados, Haiti, Nicaragua May to September 2020 period and gradually reduced to only Middle East and North Africa (MENA): Djibouti; Egypt, Arab Rep. formal employment afterward (estimates available on request). South Asia (SA): Bhutan Correlates of COVID-19 Infections Sub-Saharan Africa (SSA): Benin, Cameroon, Central African Republic, Comoros, Congo, Rep., Cote d'Ivoire, Gabon, Gambia, The, Guinea, The association between the COVID-19 infection rate and the Liberia, Madagascar, Malawi, Mauritania, Mauritius, Mozambique, policy responses is assessed using daily data for 174 countries for Nigeria, Rwanda, Senegal, Seychelles, Sierra Leone, Somalia, South the period January 1–October 31, 2020. A panel data regression is Sudan, Sudan, Zambia, Zimbabwe used with a country-fixed effects and time-fixed effects model. The results show that the introduction of both public-health and economic-support measures are positively correlated with a Countries with ongoing spread of infection (121) slower growth of COVID-19 infection cases with respect to the East Asia and Pacific (EAP): Guam, Indonesia, Japan, Malaysia, level that would have prevailed without the policy responses in a Mongolia, Myanmar, Philippines, Korea, Rep. country (figure 5). These results suggest the benefits of an Europe and Central Asia (ECA): Albania, Andorra, Austria, Azerbaijan, integrated policy approach to containment. Belarus, Belgium, Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Faero Islands, Finland, France, Conclusion Georgia, Germany, Greece, Hungary, Iceland, Ireland, Italy, This brief presents the patterns of COVID-19 infections and policy Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Luxembourg, Moldova, responses implemented by governments to contain the disease Monaco, Netherlands, Norway, Poland, Portugal, Romania, Russia, San and support citizens over time by country groups. Wide variations Marino, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, are observed in the timeline and magnitude of the infection rate Tajikistan, Turkey, Ukraine, United Kingdom, Uzbekistan and the types and intensities of policies across countries and Latin America and the Caribbean (LAC): Argentina, Aruba, Bahamas, regions. Despite the diversity in infections and policies, countries Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, Dominica, that have successfully controlled the transmission of COVID-19 Dominican Republic, Ecuador, El Salvador, Guatemala, Guyana, have imposed an intensive testing policy and implemented it effectively with high levels of testing coverage, on average. The Honduras, Jamaica, Mexico, Panama, Paraguay, Peru, Puerto Rico, econometric analysis suggests that public health policies such as Suriname, Trinidad and Tobago, United States Virgin Islands, Uruguay, testing, along with economic support, can help effectively contain Venezuela the disease, thereby providing the fundamental prerequisite for a Middle East and North Africa (MENA): Algeria, Bahrain, Iran, Islamic resumption of normalcy. Policy makers will need to strengthen the Rep., Iraq, Israel, Jordan, Kuwait, Lebanon, Libya, Morocco, Oman, health care capacity especially for early detection and testing, as Qatar, Saudi Arabia, Syria, Tunisia, United Arab Emirates, West Bank well as the capacity to reach out to the informally employed, and Gaza especially in the services sector. North America (NA): Bermuda, Canada, United States South Asia (SA): Afghanistan, Bangladesh, India, Nepal, Pakistan, Sri Appendix: Countries by Containment Group as of Lanka November 17, 2020 Sub-Saharan Africa (SSA): Angola, Botswana, Cabo Verde, Ethiopia, Least affected countries (17) Ghana, Kenya, Lesotho, Namibia, South Africa, Swaziland, Uganda East Asia and Pacific (EAP): Cambodia, Fiji, Lao PDR, Papua New Guinea, Thailand, Timor, Vietnam Europe and Central Asia (ECA): Greenland Note: The number of infection waves occurred in each country during the period of Middle East and North Africa (MENA): Yemen, Rep. January 1–November 17, 2020 were determined by the authors in reference to the country-specific infection-rate figures. The figures can be accessed and Sub-Saharan Africa (SSA): Burkina Faso, Burundi, Chad, Congo, Dem. downloaded from this URL: Rep., Mali, Niger, Tanzania, Togo https://www.dropbox.com/sh/6d19eq31jv8mo9f/AACq49NUcVG9wU5v1h_MfONXa?dl=0 References Askitas, N., K. Tatsiramos, and B. Verheyden. 2020. “Lockdown Strategies, Mobility Patterns and Covid-19.” Loayza, N. V., and C. Meza-Cuadra. 2018. “A Toolkit for Informality Scenario Analysis.” World Bank, Covid Economics 23: 263–302. Washington, DC. https://cepr.org/sites/default/files/news/CovidEconomics23.pdf Loayza, N. V., A. Sanghi, N. Shaharuddin, and L. Wuester. 2020. “Recovery from the Pandemic Crisis: Chinazzi, M., J. T. Davis, M. Ajelli, et al. 2020. “The Effect of Travel Restrictions on the Spread of the 2019 Balancing Short-Term and Long-Term Concerns.” Research & Policy Briefs No. 38, World Bank Group, Novel Coronavirus (COVID-19) Outbreak.” Science 368 (6489): 395–400. Washington, DC. http://documents1.worldbank.org/curated/en/461391599582188306/pdf/Recovery- http://documents1.worldbank.org/curated/en/461391599582188306/pdf/Recovery-from-the-Pandemic-Crisis-Balancing-Short-Term-and-Long-Term-Concerns.pdf https://doi.org/10.1126/science.aba9757 from-the-Pandemic-Crisis-Balancing-Short-Term-and-Long-Term-Concerns.pdf http://documents1.worldbank.org/curated/en/461391599582188306/pdf/Recovery-from-the-Pandemic-Crisis-Balancing-Short-Term-and-Long-Term-Concerns.pdf European CDC (Centre for Disease Prevention and Control). 2020. “Situation Updates on COVID-19.” https://ourworldindata.org/coronavirus-source-data. NTI (Nuclear Threat Initiative) and Johns Hopkins University Center for Health Security. 2019. Global Health Security Index: Building Collective Action and Accountability. Eckardt, M., K. Kappner, and N. Wolf. 2020. “Covid-19 across European Regions: The Role of Border Controls.” CEPR Discussion Paper 15178. https://cepr.org/active/publications/discussion_papers/dp.php?dpno=15178 Rae, M., and E. Friedman. 2020. “Covid-19: An Efficient and Effective Test Trace Regime Is Not a Numbers Hale, T. et al. 2020. Oxford COVID-19 Government Response Tracker. Blavatnik School of Government. Game.” BMJ 370:m3553. https://doi.org/10.1136/bmj.m3553 https://github.com/OxCGRT/covid-policy-tracker Ranna-Eliya,R., N. Wijemunige, J.R.N.A. Gunawardana, S. N. Amarasinghe, I. Sivagnanam, S. Fonseka, Y. Hasell, J., E. Mathieu, D. Beltekian, et al. 2020. “A Cross-Country Database of COVID-19 Testing.” Sci Data 7: 345. Kapuge, and C. P. Sigera. 2020. “Increased Intensity of PCR Testing Reduced Covid-19 Transmission https://doi.org/10.1038/s41597-020-00688-8. within Countries during the First Pandemic Wave.” Health Affairs December 2. https://doi.org/10.1377/hlthaff.2020.01409 Bonardi, J-P, Q. Gallea, D. Kalanoski, et al. 2020. “Fast and Local: How Lockdown Policies Affect the Spread and Severity of Covid-19.” Covid Economics 23: 325–51. Weber, E. 2020. “Which Measures Flattened the Curve in Germany.” Covid Economics 24: 205–217. https://cepr.org/sites/default/files/news/CovidEconomics23.pdf https://cepr.org/sites/default/files/news/CovidEconomics24.pdf Keita. 2020. “Air passenger mobility, travel restrictions, and the transmission of the covid-19 pandemic WEF (World Economic Forum). 2020. “The World Health Organization Has Called on Countries to ‘Test, Test, between countries.” Covid Economics 9: 77–96. https://cepr.org/sites/default/files/news/CovidEconomics9.pdf Test’ for Coronavirus—This is Why.” March 17. https://www.weforum.org/agenda/2020/03/coronavirus-covid-19-testing-disease/ Lancet Infectious Diseases, The. 2020. “Curing COVID-19” (editorial). The Lancet Infectious Diseases 20 (10): 1101. https://doi.org/10.1016/S1473-3099(20)30706-4 World Bank. 2020. World Development Indicators. 5