COVID-19, the disease caused by the novel coronavirus, has not been an equal opportunity pandemic. In the United States, the hardest-hit areas all have one feature in common: a large, nonwhite, underprivileged population, living in crowded conditions, for whom working from home is often not an option. Infectious diseases often discriminate based on income, hitting those who have less access to medical care, work jobs without sick leave, use public transport, and reside in high-density settings. But even by that standard, the coronavirus’s outsize impact on the destitute and vulnerable is striking. African Americans are already 67 percent more likely than non-Hispanic whites to be hospitalized for influenza. In the case of COVID-19, that figure jumps to 227 percent.

In the early stages of the pandemic, the virus’s devastating impact on the poor in the United States led many—including us—to predict disaster in the developing world, where poverty and overcrowding prevail and where physical distancing is rarely feasible. And yet the situation in most developing nations so far is puzzling. Most have not, for the time being, experienced the massive, nationwide outbreaks that many observers anticipated. Their public health systems have not been overwhelmed. Fears of a coming wave persist, and already several countries throughout Latin America, in the Middle East, and on the Indian subcontinent have suffered severe outbreaks and high mortality rates, but so far these are relatively isolated hot spots.

What we don’t know about the virus still exceeds what we do, but this much is clear: when it comes to COVID-19, to paraphrase former U.S. Congressman Tip O’Neill’s famous quip about politics, all epidemiology is local. How the coronavirus transmits, where it strikes most severely, and what conditions most favor its spread are questions best understood locality by locality, community by community, not at the level of continents or even countries. By the same token, even if the overall fight against the pandemic—through sharing knowledge, resources, testing kits, and, when available, a vaccine—is global, precise responses need to be local: not nationwide lockdowns, which are unsustainable, or nationwide openings, which are irresponsible, but policies tailored to specific smaller areas based on specific local knowledge. Ultimately, the fight will be won not with the public health equivalent of countrywide trench warfare but with localized guerrilla combat.


A few months into the pandemic, questions about why it has hit certain countries and areas so hard while largely sparing others far outnumber the answers. Inadequate testing means inaccurate data, and even Western countries appear to be undercounting cases by large margins. The huge gaps between death tolls in high- and lower-income countries could, in this sense, partly reflect defective information. Only when today’s deaths are measured against those occurring during similar time periods in previous years—what is commonly referred to as “excess mortality”—will the full picture emerge. Still, if hospitals in developing countries were overwhelmed and the number of funerals was skyrocketing, the world would almost certainly know. Fears of rapid spread in African nations such as Ethiopia, South Africa, or Kenya; in impoverished countries such as Venezuela or Bangladesh; or in refugee or internally displaced persons camps in Syria or Libya have not materialized to date, even if some concerning pockets of contagion have begun to emerge.

The search for explanations has experts stumped. Some argue that the virus has spared many middle- and low-income countries because their populations lack some of the biggest COVID-19 risk factors: advanced age, certain chronic medical conditions, housing in residential nursing homes, and frequent domestic and international travel. Others believe the developing world is simply experiencing a reprieve: the pandemic, barely six months old, still has plenty of time to wreak havoc in new settings.

COVID-19 has not been an equal opportunity pandemic.

Much else about the virus’s behavior continues to confound. Among the biggest surprises is the uneven geography of the outbreaks. Instead of spreading uniformly across adjoining countries, the virus seems to behave in a hyperlocalized fashion even within a country’s borders, targeting some areas and not others, with little discernible pattern to date. Neighboring countries with roughly similar demographics report strikingly different infection rates: as of Monday morning, the Dominican Republic has 184 cases per 100,000 people, whereas neighboring Haiti has only 30. Iran, with 210 cases per 100,000 people, has been hit much harder than Iraq, which has only 32. The same is true of wealthier countries, where intense outbreaks in local hot spots somehow (and contrary to earlier assumptions) do not always spread widely. Italy’s northern half, especially Lombardy, has been devastated, while the rest of the country has not. The United States and Switzerland also show huge—and so far unexplained—geographic disparities.

Another source of bewilderment: For most respiratory infectious diseases, children typically serve as vectors of infection; for instance, schoolchildren play a major role in transmitting the influenza virus. Not so in the case of COVID-19, for which current data suggest that the young may be less likely to catch or transmit the virus and that school reopenings, as have occurred in Denmark, Austria, Norway, and Finland, may not necessarily spell disaster. Once again, some exceptions stand out: in Brazil, Mexico, and India, youth seems to have provided little defense against infection. And experts are only beginning to learn about a rare but severe multisystem inflammatory syndrome that affects some children infected with the virus.


The coronavirus is not unique in affecting different people and places differently. But the extent of these variations and the difficulties in explaining them show just how much about COVID-19 still remains unknown. Filling in those gaps is crucial, for people’s economic security as much as for their physical health. If successful containment turns out to hinge on the rigor of lockdown measures, authorities would be well advised to keep some version of the lockdowns in place despite their harsh socioeconomic and political costs. If a combination of other factors proves more important, the case for maintaining lockdowns would be seriously undercut.

Lockdowns were conceived in wealthier nations, with a rationale less applicable to their poorer counterparts.

Already, lockdown measures and stay-at-home orders in the United States and Europe have drawn criticism for their alleged class blindness, with some arguing that those who can afford the draconian measures are promoting them without due regard for those who cannot. The dilemma is even more salient in poorer nations that can neither deal with the health consequences of a virulent outbreak nor cope with the socioeconomic repercussions of a long-term lockdown. Forced to strike some sort of balance between saving lives and preserving livelihoods amid scientific uncertainty, each government has gone its own way, opting for lockdowns (Uganda), a laissez-faire approach (Tanzania), or some middle point between the two.

Those who responded with strict lockdown orders have received praise for doing so, and evidence indeed suggests that countries that shut down early generally fared better than those that hesitated. At a minimum, hitting the brakes enabled officials in developing countries to assess the situation and gain time to come up with more nuanced approaches. Yet lockdowns were conceived in and for wealthier nations with a rationale arguably less applicable to their poorer counterparts. Broadly, the goal of stay-at-home policies has been to buy time to allow hospitals and other health services to prepare themselves—with ventilators and hospital beds, widespread testing capacity, and so forth—and avoid being overwhelmed by an early rush of patients. “Flattening the curve” makes sense in countries that have the means to catch up during this period. Where they don’t, where the health system already is overstretched and would take years to reach the level necessary to cope with a large-scale outbreak, the logic falters. In the absence of a vaccine or a widely available treatment, one wonders what is gained by imposing temporary lockdowns if, as soon as they are lifted (as they inevitably must be), the situation will be the same as before. 

Add to this the fact that citizens in developing countries tend to be more vulnerable to economic downturns. Their governments, meanwhile, lack the ability to inject the kinds of fiscal stimulus packages that would cushion the economic blow of shutting down. In such a setting, the death toll from economic collapse may surpass that which the virus might have caused without the lockdown. As the UN and other international organizations have warned, loss of work and wages in developing countries, especially for the vast numbers toiling in the informal economy, will translate into poverty and even famine for millions. The World Food Program has raised the specter of a possible “hunger pandemic” that could double the number of those suffering from acute hunger from 135 million to 265 million. As of now, the International Monetary Fund predicts the pandemic will push 49 million more individuals into extreme poverty. On top of this, the collateral damage to preventive health care could be severe. Vaccination, malaria prevention, treatment for chronic diseases such as tuberculosis or HIV/AIDS, and access to family-planning services are all under threat from lockdowns. Several countries have already experienced a drop in vaccination rates, putting more than 117 million children at risk of measles.

Some nations have experimented with alternative approaches. One strategy, implemented in parts of the United States and western Europe, is to shield society’s most vulnerable while allowing others to go about their lives. But the method has its weaknesses. The wide range of risk factors and the emergence of the inflammatory syndrome among some children make it hard to reliably identify susceptible individuals. Finding suitable sites to shelter those so identified and preventing contagion among them is no easier. Sweden has tried and produced questionable results, with far higher per capita deaths than its neighbors and, incidentally, a roughly similar economic toll. The de facto shielding of many elderly in the United States is another warning sign, as nursing homes have experienced some of the most serious outbreaks of COVID-19. These issues loom even larger in developing nations, where the norm is big, multigenerational families whose members depend on one another for sustenance and where quarantining or isolating the sick may not be possible.


For developing countries, crafting a sensible and sustainable policy under these circumstances is a vexing challenge. Doing so will be a work in progress, constantly subject to both minor tweaks and major change, as each day brings new data about COVID-19 and thus information on how best to address it.

Still, a few broad principles should serve as guidance. First, long-term, nationwide lockdowns are no more advisable than abrupt, indiscriminate reopenings. When lockdowns end, some activities ought to remain strictly curtailed, including large public gatherings, shared eating environments such as restaurants, and meetings in closed spaces, all of which could lead to “superspreading” events that quickly infect large numbers of people.

Second, given the virus’s geographic unpredictability, community responses are preferable to nation- or regionwide policies. The key to managing the virus in the long run is to be sensitive to local variations and adapt to them. That, in turn, requires governments to collect locally and share globally as much data as possible, which will allow them to better understand transmission factors, single out hot spots, monitor at-risk populations, and implement customized responses. That such sharing is largely facilitated by the World Health Organization makes the Trump administration’s decision to turn its back on the agency all the more irresponsible and self-defeating.

Wide-scale testing and contact tracing are prohibitively expensive and logistically challenging in most low- to middle-income nations—indeed, they are in the United States, where some 20 million tests per day might be needed to reliably monitor viral spread. A final principle, accordingly, should be to develop and roll out more affordable and accessible means of evaluating viral spread—for example, through cheaper detection kits or, alternatively, by assessing viral presence in sewage in different locations, replicating the successful reaction to polio outbreaks and enabling a surgical, area-specific response.

The virus’s unpredictability and apparent capriciousness do not mean that its patterns of transmission cannot be predicted or that it is in fact whimsical; rather, they point to large gaps in our knowledge. As countries across the globe accumulate data, they should work as de facto clinical laboratories, learning where and why the virus is striking and testing different approaches to controlling it. A painstaking task, no doubt. But it appears that the virus is here to stay and will continue to confound us. Responding with the same tenacity as it is showing seems the least that should be done.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • ROBERT MALLEY is President and CEO of the International Crisis Group and served as White House Coordinator for the Middle East, North Africa, and the Gulf States from 2015 to 2017.
  • RICHARD MALLEY is an Infectious Diseases Physician at Boston Children’s Hospital and Professor of Pediatrics at Harvard Medical School.
  • More By Robert Malley
  • More By Richard Malley