In the World Development Report 2014 - Risk and Opportunity, we put a spotlight on pandemics. There, we warned that most countries and the international community were unprepared for a risk of this nature. The world is four months into the COVID-19 pandemic, and it is now clear that we are facing an acute public health, economic, and humanitarian crisis. What makes managing this health emergency so challenging is that if unattended, it could lead to countless numbers of fatalities—yet if drastic measures to contain the spread of the virus are imposed, it can produce a deep recession with business closures, mass unemployment, and poverty. As The Economist writes, “the trade-off between saving lives and saving livelihoods is excruciating” (March 26, 2020).   

However, as Francisco Ferreira reminds us in a recent blogpost, this “trade-off is real—but its terms are not fixed in stone.” The objective of public policy in the face of the pandemic should be easing this trade-off: reducing the fatalities from COVID-19 while preserving people’s livelihoods.

In advanced countries, the lives-versus-livelihoods trade-off can be eased with immense resources. For example, the United States will spend around US$5,700 per capita to tackle the crisis, while Denmark will spend about US$7,500 per capita. Poor and developing countries face a very different situation: short of money and low in government capacity, they carry the burden of precarious health systems, overcrowded cities, and informal labor markets. 

This dire situation calls for pragmatic and effective solutions, tailored to the reality of poor and developing countries. Broadly speaking, public policy approaches to face the pandemic can be divided into two categories: 1) relief and recovery economic policies; and 2) containment and mitigation public health measures. I will address the first category only briefly here: relief and recovery measures should include a set of policies to increase public health care capacity (procuring emergency hospital space, breathing ventilators, medical protective equipment, and testing kits); protect the poor and vulnerable (scaling up both targeted and untargeted cash transfers); provide temporary support to affected businesses (granting subsidies and tax reductions); and ensure the continuity of public services. (For more on economic policy see, for instance, Berk Özler’s blogpost on social protection and my article with Steven Pennings on macroeconomic policy). I will devote the rest of this blogpost to containment and mitigation measures. 

In the 2014 World Development Report, we propose some principles for effective risk management that form the basis for the recommendations that follow. The first is that preparation is the key to resilience, even during a crisis. The second is that coping with adverse systemic shocks requires a whole-of-government, whole-of-society integrated response. The third is that public actions should consider the realities of each country, making sure that the unintended costs do not outweigh the potential benefits.

With poor and developing countries in mind, let me share some thoughts on what not to do and what to do regarding containment and mitigation measures. 

The Problems with Lockdowns

I have serious doubts about the efficacy of lockdowns. I recognize that governments have had difficult choices to make on the best approach for their countries to contain the spread of the disease. Surrounded by uncertainty as to the threat of the virus and the prevalence of the infection in their countries, some governments chose lockdowns. With the benefit of more evidence and time, I believe we can make better choices. 

Using clear-headed cost-benefit analysis, Julian Jamison concludes that “We should practice sustained yet moderate social distancing—but not a lockdown” (The Incidental Economist, April 1, 2020). Jamison’s conclusion applies to advanced economies, but they are even more relevant to poor and developing countries where lockdowns tend to be indiscriminate and disorganized. In Ideas for India, Debraj Ray and S. Subramanian argue against a comprehensive lockdown and in favor of a “reasonable alternative” whereby young adults are allowed to work and the elderly are insulated and cared for in their own households. Such measures can be guided by antibody testing when possible (March 28, 2020). 

Let me summarize why I think lockdowns are ineffective and excessively costly, especially in developing countries. 

Lockdowns are ineffective in containing the spread of the disease when: 

• They are imposed in cities with pervasive overcrowded dwellings and neighborhoods. There, instead of social distancing, the result from a lockdown is social compression: “Confining people in cities such as Lagos, Mumbai, or Manila where half of the population may live in slums, will force them to be crammed in a room with six or eight people without easy access to water or soap” (The Financial Times, April 2, 2020).
• They produce massive displacement of people, especially from urban to rural areas, spreading, rather than containing, the contagion of the virus. This is happening in India and Kenya, for instance, where migrant workers who depend on daily labor are escaping cities in lockdown in massive numbers and returning to their hometowns (NPR, March 31, 2020).

And lockdowns are excessively costly in economic and human terms because: 

• They can lead to mass unemployment and bankruptcies. Most developing-country governments do not have the means to support people and businesses in a deep recession. They have large levels of external debt, low tax revenues, and high credit risk—a situation worsened by the pandemic-related drop in commodity prices  (Research & Policy Briefs, March 26, 2020). 
• They can put the families of informal workers, especially daily laborers, at the risk of starvation, crime, and disease. From South Asia to Africa to Latin America, hundreds of millions of informal workers, without unemployment insurance, paid leave, or savings, would rather work and face the risk of infection than starve (The Wall Street Journal, April 2, 2020). 
• They may create the conditions for unbridled domestic violence. Being in close quarters with an abusive spouse or parent is dreadful. It can be even worse when police protection is unavailable as police resources are diverted to enforcing the lockdown. In Hubei province in China, the initial epicenter of the coronavirus outbreak, official reports of domestic violence tripled during the quarantine. “Women and children who live with domestic violence have no escape from their abusers during quarantine, and from Brazil to Germany, Italy to China, activists and survivors say they are already seeing an alarming rise in abuse” (The Guardian, March 28, 2020).
• They can be manipulated to suppress political opposition and legitimate social unrest in authoritarian and dictatorial regimes. In several countries, governments have declared states of emergency and deployed the armed forces: “The greater worries lie elsewhere, in the abuse of office and the threats to freedom. Some politicians are already making power grabs…” (The Economist, March 26, 2020).
• The opportunity cost of the loss of public and private resources due to lockdowns is immense in poor and developing countries. This loss implies, for instance, a diminished ability to provide other vital services in health care, education, and safety. This, added to the loss of jobs and income, can significantly worsen poverty and vulnerability (World Bank EAP Update 2020).

Worse than a lockdown is a series of repeated and uncertain lockdowns. The risk of second and third waves of infection is large in populations with low immunity. The prospect of repeated and uncertain lockdowns can devastate the economy and worsen human suffering beyond comprehension. Whether or not they have tried lockdowns as a first line of defense, countries around the world, especially poor and developing countries, should turn instead to sustainable smart containment and mitigation measures.

Smart Containment and Mitigation Measures

Although there is much uncertainty regarding the science around COVID-19, including its epidemiology, there is some evidence that can help guide smart containment and mitigation measures. (For the evidence, see Center for Disease Control 2020; and for a summary of health infrastructure for containment, see Box I.B.5 in World Bank EAP Update 2020). Smart containment and mitigation measures are especially relevant where lockdowns are ineffective and excessively costly. They are also important for countries that are considering sensible exit strategies from their lockdowns to prevent a second wave of infection (see Piero Ghezzi´s “Ideas for an Exit Strategy”). Some of these actions may seem simple, but they are not always easy to implement in poor-country contexts. 

1. Practice effective personal and public hygiene. Thoroughly washing hands with soap and water or using a hand sanitizer that is strong enough is a powerful defense. Yet, in many poor countries, people don’t have access to water (2.2 billion people, according to WHO 2019) and cannot afford to buy enough soap and sanitizer. Consider the following recommendations:
   • Conduct massive, imaginative, and persuasive public campaigns for people to wash and clean their hands properly. 
   • Install public hand-washing stations where water is scarce. This was done in West Africa during the Ebola crisis, installing simple devices (two buckets and some chlorine) in public buildings and schools (NPR, March 30, 2020).
   • Install and maintain sanitizer dispensers in public places, such as schools, public transport, restaurants, banks, and shops. 
   • Disinfect public places frequently and extensively.  
2. Protect the most vulnerable to the disease. The elderly (and people with certain underlying health conditions) seem to be the most vulnerable to COVID-19. In theory, we would want to insulate the elderly while enabling the young to keep the economy going. In practice, cocooning the elderly may be a high-risk and socially unacceptable strategy. However, some forms of less drastic insulation may be feasible. (Look for a forthcoming piece by Adam Wagstaff on protecting the elderly.) Consider the following recommendations:
   • Household-level shielding for the elderly (and possible other vulnerable people). Incentives can be provided for the elderly to stay at home and their family members to practice safe behaviors around them. Dahab et al. (2020) offer some general principles and practical procedures to shield high-risk populations. The government can help by providing information to the families and cash transfers to the elderly as support and incentive for suitable behavior. 
   • Social shelters for the elderly. This can be designed for families that cannot provide for the safety or the insulation of their elderly relatives. Providing these shelters is a challenging task for any government, especially for those with low capacity. The international community, including charitable organizations, could help with this responsibility. Ideally, these shelters should be run by people who have acquired immunity to the coronavirus (which, of course, implies antibody testing). 
3. Isolate people who are currently infected. Targeting, tracing, and monitoring those who are currently infected and those who may have been exposed to the coronavirus are best practices that have allowed South Korea and Japan to control the outbreak while avoiding lockdowns (The Economist, April 3, 2020; and ). Proper testing is needed to make targeting, monitoring, and case tracing feasible (see point 4). While testing is procured, other complementary measures can be implemented. Consider the following recommendations, 
   • Self-quarantine. With the information provided by antigen testing, residents and international travelers can be ordered to self-quarantine. In some cases, it may be necessary for the government to provide the incentive and support of a cash transfer. 
   • Social-quarantine. If household conditions are not propitious for self-quarantine, the government can consider providing sheltering facilities for people who are infected but are not in need of critical medical care. They would also get a cash transfer, possibly to help support their families. Ideally, these places for social-quarantine would be run by people who have acquired immunity to the coronavirus. 
   • Employ digital data. Enforcing self- and social-quarantine can be facilitated by the use of digital data devices, such as mobile phone apps. They can also help in tracing contacts and in providing information on coronavirus hotspots. 
   • Require face masks in public. The government can procure large supplies of masks, distribute them for free, and make their use mandatory in certain public places, such as public transport, shops, and markets. This is important because a large portion of individuals carrying the coronavirus are asymptomatic. Since the primary purpose of wearing masks is protecting others, social enforcement can make this practice feasible and prevalent. After initial reluctance, the US Center for Disease Control now “recommends wearing cloth face coverings in public settings where other social distancing measures are difficult to maintain (e.g., grocery stores and pharmacies)”; and some countries in Europe, including Austria, have made this practice mandatory (Financial Times, April 2, 2010). 
4. Test, test, test. Mass testing for the coronavirus is an indispensable piece of information for smart containment and mitigation strategies. (Look for a forthcoming piece by World Bank economists Damien de Walque, Jed Friedman, and Aaditya Mattoo on COVID-19 testing.) The antigen test signals who is currently infected by the coronavirus and is likely contagious to others. The antibody test indicates who has been exposed to the coronavirus in the past and is likely to have developed immunity to it. Reliable, rapid, and scalable tests are currently being developed, with promising signals. Right now, the bottleneck for mass antigen and antibody testing is technological in advanced countries. Once this bottleneck is resolved, the challenge for developing countries will be large-scale implementation. De Walque, Friedman, and Mattoo recommend starting with targeted testing in the near term and, as capacity expands, scaling up testing to comprehensive levels. Consider the following recommendations:
   • Antigen testing. Apply the antigen test to susceptible groups to determine who should be quarantined, who should be traced for potential infection (contact tracing), and what areas could be treated to prevent contagion hotspots. Since this test is available, obtain antigen testing kits from international companies and agencies. International organizations and aid agencies could provide knowledge, materials, and funding to obtain them.     
   • Antibody testing. When this becomes available, apply the antibody test first in a representative sample to determine the acquired immunity profile of the population (by area, age, gender, and other salient characteristics) and whether or not “herd” immunity is likely to have built up (The Economist, March 2, 2020). Then, scale up the test to obtain individual information to guide decisions on work and employment, access to public services, and contact with vulnerable groups. Enlist the help of international organizations to deploy and implement the test.
   • Implementation capacity. Prepare and improve the implementation capacity of the public health care system to administer the new antigen and antibody tests, as soon as they are developed. This can involve enlisting the help of medical students and other university students, for instance.

The benefits of smart containment and mitigation efforts can be substantial. In their theoretical study of the macroeconomics of epidemics, Eichenbaum, Rebelo, and Trabandt (2020) analyze the gains of conditioning containment policies on people’s health status, so that infected people would not work, susceptible people would work but less than they normally do, and recovered people would work more. This “smart” containment policy significantly eases the lives-versus-livelihoods trade-off: it decreases the infection rate to less than half and the severity of the recession by more than 5 percentage points. 

In practice, to approach and improve over this scenario, we need the knowledge that infection and immunity tests can provide, the ability to isolate people who are infected, the capacity to shield vulnerable populations, and a set of reasonable measures to diminish contagion. While the gains of smart containment and mitigation measures could be large in advanced countries, for developing countries they could literally be a matter of life and death.
 

_{For ideas and suggestions, I am very grateful to Susmita Dasgupta, Damien del Walque, Sharmila Devadas, Jed Friedman, Piero Ghezzi, Olga Jonas, Aart Kraay, David Mackenzie, Aaditya Mattoo, Berk Özler, Steven Pennings, Firas Raad, Nurlina Shaharuddin, Mike Toman, and Adam Wagstaff. They are, however, not responsible for the opinions expressed in this note, and neither are the World Bank, its Executive Directors, or the countries they represent.}