Questions and answers about
the economy.

Past plagues: why did some create conflict while others promoted cohesion?

History provides a reservoir of accumulated knowledge for coping with a deadly disease like Covid-19. Importantly for policy, it offers insights into what determines whether a society tends towards greater violence or greater cooperation in the aftermath of a pandemic.

The Covid-19 pandemic and its fallout have taken the world by surprise. Yet the history of pandemics and epidemics offers an essential window into how the spread of pathogens can shape society and human behaviour. As the coronavirus outbreak has unfolded, and particularly given emerging evidence of animus and discrimination against certain groups accused of causing or spreading the disease, there has been renewed interest in understanding the social effects of past large-scale outbreaks.

Some historical episodes gave rise to violence and the ‘scapegoating’ of minority groups, while others did not. In fact, some even improved social cohesion and increased cooperation. What trajectory a society takes after major disease outbreaks depends on first, the epidemiological characteristics and medical knowledge of the disease; and second, the prevailing social, economic and political context.

Evidence from both historical pandemics and more recent epidemics can shed light on which policy levers can be used to prevent social conflict from arising during such events. Given the rare nature of these events in the past, a broad coverage of historical outbreaks is needed to establish any patterns. This requires compiling the evidence from a wide range of studies covering events from the pre-industrial era (such as the plagues of antiquity and the Black Death) to industrial-era outbreaks (such as cholera, the influenza pandemic of 1918-19 commonly known as the Spanish flu, and more recent experiences with HIV and Ebola).

The study of such historical evidence, in conjunction with economic theory, provides insights into the conditions and factors that determine how social cohesion is affected in the aftermath of pandemics and epidemics.

When do pandemics lead to conflict and scapegoating?

Studies from political science, psychology and sociology posit scapegoating as a possible cause of violence in times of high distress (Doob et al, 1939; Hovland and Sears, 1940; Girard, 1978; Allport, 1979; Staub, 1992; Poppe, 2001; Glick, 2002, 2005, 2009). Members of a majority group experiencing prolonged negative experiences settle on a specific target to blame for their grievances and may experience emotional relief by blaming a minority group.

Many studies document that past pandemics resulted in the scapegoating of minorities (for example, Nelkin and Gilman, 1988; Eamon, 1998; Craddock, 2004; Cohn, 2012, 2017, 2018). The Black Death of 1347-52 led to the mass persecution of Jews, and the cholera, smallpox and plague riots of the 19th century and the Ebola outbreaks of the 20th century led to violent scapegoating, which resulted in murderous attacks against health and governments officials.

Other outbreaks – plague recurrences after the Black Death, syphilis, the Spanish flu and HIV – led to milder scapegoating but did not necessarily cause large-scale and systematic social violence. Rather, minority groups were blamed for disease outbreaks, which led to cases of medicalised prejudice, discrimination and individual cases of targeted violence. There is also evidence of increased discrimination against migrants, for example, against African migrants during the HIV crisis (for example, Edelstein et al, 2014) and the Ebola crisis (for example, Lin et al, 2015).

Today, media reports have suggested an increase in discrimination against people of Asian descent in the United States (New York Times, 2020, Washington Post, 2020) and the rest of the world (Human Rights Watch, 2020) given that the origins of the virus were in China. Discrimination against minority religious groups in South Asia (Human Rights Watch, 2020; The Guardian, 2020a; Al Jazeera, 2020) and against people of African origin in Chinese cities (The Guardian, 2020b) has also been cited.

Migrants in general have been blamed for being ‘super-spreaders’ given that they travel significant distances to find work, are more likely to become infected and are at a higher risk of being carriers of infection when they return (Ahsan et al, 2020; Khanna et al, 2020).

But such scapegoating is not inevitable. There are also historical instances, such as the plagues of antiquity and the yellow fever outbreaks of the 18th and 19th centuries, which did not particularly lead to any blame or violence. In fact, in some cases, they generated greater compassion and increased social cohesion among the survivors. Historical records and media may tend to focus excessively on the most extreme examples of violence and minority persecution during epidemic times due to ‘selection bias’ (Cohn, 2012).

What then determines whether a society tends towards conflict or cohesion in the aftermath of a pandemic event? Of particular importance are the epidemiological characteristics of the disease in question and the social, economic and political context of where and when the outbreak occurs.

These are contingent on the prevailing state of scientific knowledge, available technologies and the information (or misinformation) that is disseminated about the disease. Together, these determine individuals’ wellbeing and behavioural responses during the pandemic, and the collective responses of social groups.

Why the epidemiology of a disease matters

The epidemiology of a disease determines how it spreads and who it kills. Its biological characteristics – such as its contagiousness, the mode by which it spreads, the visibility of its symptoms, the prevalent understanding of disease, and case fatality rates – all determine how deadly and distressing a virus is. The deadlier a disease, the more emotional distress it may generate, and the likelier it is to result in violent scapegoating.

The symptoms of the Black Death were particularly frightening – large black buboes would form all over the body and extremities would turn black – and at 70%, its case fatality rate was very high. Commentators suggest that it was the sheer scale and devastating impact of the Black Death that made contemporaries feel that it was either the wrath of God or part of a grand conspiracy (Nohl, 1924; Horrox, 1994). This lack of medical understanding of the disease eventually led to the plague being attributed to the poisoning of wells by Jews, which led to mass expulsions and murders.

It may also matter who is most vulnerable to the disease. Social violence could be more likely when a disease kills working-age adults because they are the ones who commit the persecutions. Violence may be more likely if a disease disproportionately harms the poor (for example, if poor living conditions such as congested housing and poor sanitation increase susceptibility), since they may be more able to form mobs (since the poor are more numerous). Historical evidence also suggests that pandemics that kill children more are the most socially dangerous as they may trigger the strongest emotional responses.

How social, political and economic contexts matter

The prevalence of inequality and intergroup dynamics also have important roles in determining whether pandemics lead to social unrest. Studies suggest that inequality between groups can lead to conflict if it causes grievances or provides incentives for a relatively poor group to engage in conflict (Kanbur, 2007; Blattman and Miguel, 2010). Grievances can be exploited to obtain gains by violent means, and ethnic identity can serve as ‘a strategic basis for coalitions that seek a larger share of economic or political power’ (Ray and Esteban, 2017).

Jews, for example, often specialised in medicine during Medieval times, which is likely to have given rise to suspicions about their knowledge of poisons. At the same time, however, Jews also served as moneylenders and tax collectors, thus resentments against them possibly existed prior to the Black Death and were only inflamed by it. Similarly, reports of discrimination against minority groups and immigrants in the aftermath of the Covid-19 come against the backdrop of a global increase in populist sentiments that predates the pandemic.

At the same time, minority groups may provide specialised skills that can help a society to eliminate an epidemic or recover faster after it. In cases like that, policy-makers may provide protection to such groups. For example, Jews in Medieval Europe played a vital role in relatively high-skilled occupations such as money lending and trade, and the economic value they generated was clearly recognised at the time (Chazan, 2010; Johnson and Koyama, 2019). Studies suggest that many cities provided protection to their Jewish quarters against mob violence, especially where they held important economic roles (Rowan, 1984; Wasserman, 2007; Jedwab et al, 2019).

It is not necessary that scapegoating be restricted to religious or ethnic groups. Violent ‘cholera riots’ took place in many cities of various industrialising nations throughout the 19th century (Cohn, 2012, 2017). Cholera disproportionately killed the urban poor in congested 19th century industrial cities in a time that was already dominated by a constant and violent class struggle between the bourgeoisie and the proletariat. In this context, the population believed ‘elites, with physicians as their agents, had invented the disease to cull populations of the poor’ (Cohn, 2018).

The role of scientific knowledge and information or misinformation

Experience from past pandemics also points to the role of medical understanding and beliefs about diseases in how societies responded. Until the late 19th century, ‘bad air’ due to the weather, the misalignment of planets or unsanitary overcrowding were often deemed responsible for sickness. As such, specific groups were less likely to be blamed.

Poison then became a major form of killing in the Medieval and Renaissance periods (Wexler, 2017), as the Islamic Golden Age (of the 8th to 14th centuries) contributed to major advances in pharmacology (Hadzovic, 1997). It is thus not surprising that epidemics during this period became increasingly associated with accusations of poisoning.

Even as technology and scientific understanding improved, other social cleavages, such as inequalities in education and access to new technologies, gave rise to social conflict. For example, after the germ theory of disease became more established by the 1850s, some epidemics led to scapegoating against disease victims.

In the case of the smallpox epidemics of the 19th century, a vaccine already existed in 1796 (Wolfe and Sharp, 2002). The poor, who in large part became victims of smallpox, were seen by the elites as ‘guilty’ for succumbing to their infection, and were blamed for their ignorance or lack of consideration for the rest of society. Likewise, when a treatment became available for syphilis in 1910, women who contracted syphilis also came to be seen as ‘guilty’.

Lastly, one may wonder whether there will be more social conflict associated with pandemics in the future, given the increasingly lower costs of spreading falsehoods. Today, mostly due to the rise of the internet, there are many channels of information. As a result, false information is more widely propagated even when it is scientifically untrue and the statements are falsifiable and have been falsified.

There are then political (and economic) rewards for those engaged in falsification. Combined with the lower costs of spreading disinformation today, falsification may continue to trend upward, including in a coordinated manner between individuals seeking the same rewards.


The many studies of the effects of epidemics on social cohesion can help to inform the policy debate on the effects of, and best responses, to Covid-19, especially for developing countries. Recent epidemics have featured either low rates of contagion (for example, Ebola) or relatively low fatality rates (for example, Covid-19).

But what if a pandemic that is both highly contagious and has a high fatality rate were to spread? A deadlier disease outbreak may bring a higher risk of violence. For example, given its high fatality rate (50%), an uncontrolled Ebola pandemic could have dramatic effects (United Nations Economic Commission for Africa, 2015; Gates, 2018).

This stresses the importance of studying the effects of past epidemics and pandemics, and highlights the fundamental role that history plays as a reservoir of accumulated knowledge to safeguard and tackle the next, perhaps, deadlier pandemic.

Authors: Remi Jedwab, Amjad Muhammad Khan, Richard Damania, Jason Russ and Esha Dilip Zaveri
Note: We thank Peter Lanjouw for helpful comments. This project was gratefully funded by the World Bank’s Water Global Practice.
Photo by Hamilton Henry Dobbin via Wikimedia Commons
Recent Questions
View all articles
Do you have a question surrounding any of these topics? Or are you an economist and have an answer?
Ask a Question
Submit Evidence