key: cord-0841628-wb75s6ey authors: Ide, Tobias title: COVID-19 and armed conflict date: 2020-12-14 journal: World Dev DOI: 10.1016/j.worlddev.2020.105355 sha: 9ab6f624e1501bcee815aeae8c008fddbff1d8fe doc_id: 841628 cord_uid: wb75s6ey This article studies the impact of COVID-19 on armed conflict. The pandemic has significant health, economic and political effects. These can change the grievances and opportunity structures relevant for armed conflicts to either increase or decrease conflict risks. I analyse empirical evidence from Afghanistan, Colombia, India, Iraq, Libya, Pakistan, the Philippines, Thailand and Yemen from the first six months of 2020. Results suggest that COVID-19 provides little opportunities for health diplomacy and cooperation, but it also has not yet driven grievances to a level where they became relevant for armed conflicts. Four countries have encountered temporary declines in armed conflicts, mostly due to strategic decisions by governments or rebels to account for impeded logistics and to increase their popular support. Armed conflict levels have increased in five countries, with conflict parties exploiting either state weakness or a lack of (international) attention due to the COVID-19 pandemic. This is a worrisome trend given the tremendous impacts of armed conflict on human security and on the capabilities of countries to deal with health emergencies. COVID-19 is an infectious disease caused by the SARS-CoV-2 virus. At the day of writing (November 11, 2020) , at least 51 million people worldwide have been infected with COVID-19, and more than 1.2 million have died. Government measures to combat the spread of SARS-CoV-2 have resulted in various restrictions and lockdowns (Ritchie et al., 2020) . The economic consequences of this are devastating. The most recent World Economic Outlook diagnosed a real GDP growth of À4.4% for 2020 -the deepest recession since 1945 (IMF, 2020) . Currently, scholars and policy makers assign high priority to limiting the spread of the virus, improving treatment, developing a vaccine, and mitigating adverse economic effects. Contributing to these efforts, I argue that it is also of utmost importance to monitor and analyse the impact of COVID-19 on armed conflict, with an emphasis on its most prevalent form, intrastate armed conflict. Outbursts of physical violence involving at least one organised group claimed on average more than 105,000 lives per annum for the last six years (Pettersson & Öberg, 2020) . Armed conflict therefore has enormous negative effects on human security and effectively constitutes ''development in reverse" (Collier, 2007) . Past studies indicate that contagious diseases have a tangible impact on the onset and the duration of armed conflicts (e.g., Cervellati, Sunde, & Valmori, 2017; Kustra, 2017) . Factors like GDP decline and poor health are also among the established predictors of civil war (Cederman & Weidmann, 2017) . UN Secretary-General António Guterres therefore concluded that ''the pandemic also poses a significant threat to the maintenance of international peace and security" and called for ceasefires to aid responses to COVID-19 (Clarke, 2020) . An increasing frequency or intensity of armed conflict, in turn, would seriously affect efforts to cope with the pandemic (Mehrl & Thurner, 2020) . Armed groups collaterally or intentionally destroy infrastructure that is essential to a healthy life, such as water, electricity or health care (Sowers, Weinthal, & Zawahri, 2017; Wise & Barry, 2017) . In line with this, Wagner et al. (2019) find that high-intensity armed conflict increased the mortality of female non-combatants in Africa by 202%. In the worst case, migration flows triggered by violence may spread SARS-CoV-2 even further. In this article, I theorise how COVID-19 can affect armed conflict risks (Section 2). I then analyse armed conflict trends and their drivers in nine countries in the period January to June 2020 based on quantitative and qualitative evidence (Section 3) before drawing a conclusion (Section 4). Armed conflicts are complex, dynamic and multi-faceted phenomena. Their onset, intensity and duration is driven by a wide range of factors (Cederman & Vogt, 2017) . This results in a limited explanatory power of single variables like COVID-19. However, based on established theoretical frameworks and previous research, there are good reasons to assume that COVID-19, the associated restrictions and their economic fallout affect armed conflict dynamics. Broadly speaking, armed conflicts can be driven by grievances and opportunities (Taydas, Enia, & James, 2011) . Grievances refer to intense perceptions of inequality or unfairness by individuals or social groups. If the latter see no perspective for addressing such unequal or unfair situations in a non-violent manner, they can resort to physical violence. However, Fearon and Laitin (2003) argue that grievances are far too widespread to explain relatively rare phenomena like armed conflict. In their view, explanations should instead focus on factors that provide aggrieved groups with the opportunity to start and sustain violent confrontations, such as the availability of recruits, weapons, and hideouts. In practice, the grievances and opportunity perspective are often deeply intertwined, for instance when aggrieved individuals can be more easily recruited by armed groups. COVID-19 can affect grievances to an extent that makes armed conflict both more and less likely. Many consequences of the pandemic such as poor health, poverty and economic recession can lead to significant individual frustration, and are well-established predictors of armed conflict (Cederman & Vogt, 2017) . Strained budgets and additional financial demands reduce the ability of the state to appease such grievances (Price-Smith, 2009). Lockdown measures are themselves a significant point of contention as indicated by protests around the world that oppose them (Reuters, 2020) . Furthermore, the Armed Conflict Location and Event Dataset (ACLED) records a significant number of protests and attacks against ethnic and religious minorities suspected of transmitting SARS-CoV-2 (Raleigh, Linke, Hegre, & Karlsen, 2010) . Such xenophobic attributes and actions harden pre-existing cleavages. COVID-19 might also provide a chance to demonstrate solidarity and good intentions, and hence lessen grievances. The literature on health diplomacy, for example, discusses how cooperation on shared health challenges can increase the prospects for peaceful relations. The empirical success of such efforts is so far been limited (Kelman, 2019) . However, research on environmental peacebuilding has revealed that low-level, mutually beneficial cooperation can yield peace dividends in certain contexts (Ide, 2019) . Furthermore, ceasefires to deliver health benefits have at least temporally reduced armed conflict intensity on several occasions in the past (Chattu & Knight, 2019) . In response to the pandemic (and António Guterres' call), armed groups in 14 countries have announced ceasefires to support responses to COVID-19 (Rustad, 2020) . The COVID-19 crisis can affect opportunity factors to increase armed conflict risks. As GDPs decline, unemployment is on the rise and international remittance flows are reduced by around 20% (World Bank, 2020) , livelihood insecurity will grow. This results in lower opportunity costs for individuals joining an armed group vis-à-vis seeking legal employment, hence facilitating recruitment by violence entrepreneurs. In the same vein, Kustra (2017) argues that high levels of disease prevalence and the associated loss in life expectancy reduces the relative risks of individuals for joining dangerous activities like rebellion. Capable states can deal with these impacts of COVID-19 by extending social security nets, mediating emerging conflicts, and disarming violent groups (Sobek, 2010) . But COVID-19 also undermines state capability: While financial demands to the state growth, its fiscal base is strained due to a loss World Development 140 (2021) 105355 of tax revenues. The collapse of tourism and primary commodity prices (such as oil) affect the income of many states as well. Furthermore, members of the police and military might get infected, or are re-deployed to assist measures to contain the disease (Ataguba, 2020; Bagozzi, 2016) . However, COVID-19 might also shape opportunity costs in a way to reduce armed conflict risks, at least temporarily. If a state's capability is strained and there is an urgent need to deal with a health emergency, military offensives are certainly unlikely (Price-Smith, 2009 ). Furthermore, existing as well as potential rebel groups and militias face similar challenges in the face of the pandemic. They need to raise money and food to supply to their fighters during an economic recession, convince their members to take part in operations rather than staying at home (to reduce infection risks and support their family or community), and deal with the logistical constraints of lockdowns and border closures. Starting or intensifying attacks during the COVID-19 crisis is likely to decrease the local (and international) legitimacy of armed groups, especially if health infrastructure is affected. The ceasefire declarations by armed conflict parties in several countries can also be interpreted as a sign that COVID-related capability and legitimacy concerns are warranted. Analysing the impact of COVID-19 on armed conflict is difficult at this stage of the pandemic. After many countries were initially able to flatten infection curves and ease restrictions, new peaks of infections and lockdowns are currently seen. The long-term economic, political and health impacts of the pandemic are hard to predict. At the same time, COVID-related data are subject to uncertainties and political manipulation (Linsi & Aragão, 2020) . This understanding is pertinent to my analysis of armed conflict dynamics in the face of COVID-19 Here, I analyse quantitative and qualitative evidence for nine countries that experienced significant levels of armed conflict when the pandemic began to unfold in March 2020. These countries reflect a variety of geographical, political and economic contexts as well as conflict types, but were also chosen due to data availability concerns. Sub-Saharan Africa is not covered by my sample (Desmidt & Neat, 2020) . Fig. 1 visualises the total number of battles and explosion events between January 1, 2019 and June 30, 2020 in the countries under study. As shown by Fig. 2 , eight of the nine countries experienced a rapid spike of COVID-19 infections starting in mid to late March, while infections in Yemen only rose in late April. Until the end of June 2020, all countries under study were able to bend their infection curves, hence making them broadly comparable. It is important to recognise here that while some factors connecting the Corona crisis to armed conflict depend on the number of infections (e.g., infected members of the military and armed groups, higher mortality rates), this is not necessarily the case for many others (e.g., restrictions, economic recession). Fig. 3 illustrates the strictness of government responses to COVID-19 in the countries under study over time. It draws on the government response stringency index which summarises policy responses like school and workplace closures, the cancellation of public events, restrictions on public gatherings as well as limitations to internal and international movement. All countries started with low to moderate restrictions in March, accelerating in April and May to index values above 80, which indicate very tight restrictions (with the exceptions of Thailand and Yemen, which also saw the lowest case numbers). In June, governments eased restrictions in the face of plateauing case numbers (Yemen is again an exception). Overall, four countries experienced a remarkable decline in armed conflict events (battles and explosions) during the first months of the COVID crisis. The downward trend in Afghanistan is in part due to the February 2020 peace deal between the USA and the Taliban. But Fig. 2. . Confirmed COVID-19 cases per day per one million inhabitants (real numbers are likely to be higher). Source: Ritchie et al. (2020) . World Development 140 (2021) 105355 COVID-19 has also strained the capability of state security forces to carry out attacks, for example because soldiers had to support the pandemic response. Infections also rose among Taliban fighters. The Taliban have nevertheless rejected government calls for a ceasefire. Since late March, the group has also deployed personnel to deal with COVID-19 (rather than to its spring offensive), including public information campaigns, distribution of goods, and enforcing quarantining measures. This move is strategically motivated and envisioned to serve as a ''ladder" that helps the group to gain public support and eventually oust the Afghan government (D'Souza, 2020; Jackson, 2020; Kapur & Saxena, 2020) . The armed conflict in Colombia between the government and the Ejército de Liberación Nacional (ELN) de-escalated considerably after the pandemic struck the country. On March 30, the ELN declared a ceasefire to ease coping with COVID-19. While humanitarian rather than strategic considerations drove this decision, the ELN seems less committed to permanent peace negotiations, and did not extend the ceasefire beyond May 1. The Colombian government and military did not commit to a ceasefire at all, but instigated fewer attacks. However, this is likely because resources were shifted to support the pandemic response. This is considering that the military was heavily involved in response efforts and that Colombia has the highest number of infections per capita in my sample. There are reports, however, that increased poverty and school closures due to the pandemic facilitate recruitment by the ELN, which would increase conflict risks over the long term (Alsema, 2020; Burnyeat & Gomez-Suarez, 2020; Loaiza, 2020) . In Thailand, the intensity of the conflict between the separatist Barisan Revolusi Nasional (BRN) and the government declined from April onwards. On April 3, the BRN declared a cessation of armed activities to help inhabitants in its areas of operation dealing with COVID-19. This ceasefire was predominately strategically motivated. The BRN's operative capabilities were weakened by internal travel restrictions, a lack of retreatment areas due to the closing of the border to Malaysia, and Thai military offensives in early 2020. The groups' leadership was also concerned that attacks in the wake of the pandemic (and Ramadan) would reduce its support among the local population. The Thai government did not reciprocate the ceasefire. When infections declined in late April, it resumed the war against the BRN, while the latter still lacked the capabilities for a full-scale response (Keating, 2020; Pathan, 2020) . Yemen was hit relatively late, yet hard by COVID-19. 1 On April 9, the coalition supporting the internationally recognised Yemenite government led by Saudi Arabi declared a unilateral ceasefire to support the pandemic response, but the ceasefire only lasted one day. Health diplomacy has failed in this case. However, the ceasefire seems to reflect a reduced willingness of the Saudi Arabian government to commit resources to the Yemeni civil war in the face of its own COVID-19-related problems. Likewise, the other main party to the conflict, the Houthi rebels, had to devote more resources to manage the pandemic and deal with the associated grievances in the areas it controlled, while support from Iran (heavily affected by COVID-19 and US sanctions) declined. These diminished capabilities of both conflict parties caused a strong reduction of fighting activities from late April onwards (Michael, 2020; Mustasilta, 2020) . At the same time, five countries experienced an increase in armed conflict activity during the first months of the pandemic. The growing number of armed conflict events in India was not related to the Maoist insurgency. The presence of state security forces on the ground has been reduced due to fears of infection and the Maoists' supply lines were negatively affected by a comprehensive lockdown. There are concerns, however, that the rebels use the lack of state presence and economic deprivation caused by a heavy lockdown to recruit for future offensives (Bhardwaj, 2020; Kujur, 2020) . Armed confrontations in the Kashmir region contested between India and Pakistan, by contrast, increased significantly. Clashes between both countries' militaries were a result of longer-standing tensions and thus unrelated to the pandemic (Staniland, 2020) . There is however evidence that Pakistan's support for pro-Pakistani insurgents increased to put additional pressure on India during the COVID-19 crisis. At the same time, the Indian army capitalised on the comprehensive restrictions and the occupation of public attention with the pandemic to launch a heavy crackdown campaign against (presumed) insurgents in Kashmir (Basu & Philip, 2020; Islam & Uzair, 2020) . Communal tensions in India also rose because of disputes related to permits for and infections linked to Hindu and Muslim religious gatherings during the pandemic. So far, these tensions rarely translated into larger violent confrontations (Kapur, 2020) . In Iraq, the capabilities of the government have been severely strained by the crisis, among others because oil prices collapsed and military forces were preoccupied with COVID-19 responses (e.g., enforcing curfews). Due to the pandemic, the international coalition supporting the government has also stopped training activities and some joint missions, and pulled out troops. The Islamic State (IS) was affected financially by the crisis as well due to its involvement in oil trade and the general economic decline. Nonetheless, the group sought to exploit the current weakness of the Iraqi state to expand its territorial control, thus launching additional attacks. The rise of violence in Iraq during the first months of the pandemic has been modest and non-linear (perhaps due to Ramadan in late April and May), but there was a clear upward trend of IS-initiated attacks (Hanna, 2020; Sattar, 2020) . The civil war in Libya between the Government of National Accord (GNA) and the Libyan National Army (LNA) has intensified since March 2020. Both parties aimed to launch decisive strikes and received significant logistical and material support from their international patrons. Therefore, an escalation of the war would have taken place irrespective of COVID-19. But the pandemic accelerated this escalation in two minor ways: It distracted the world's attention from the fact that both sides ignored the peace agreement concluded in January 2020. Furthermore, the GNA and the LNA believed that the other side might collapse very soon under the combined pressure of military offensives and the virus (Allahoum, 2020; Mustasilta, 2020) . Conflict intensity saw a rapid decline in June 2020, but this was due to the mutual acceptance of a military stand-off and renewed peace negotiations, rather than related to the pandemic (Gosh, 2020) . The military of Pakistan engaged in more battles with the Indian army and local Taliban groups from April 2020 onwards. The intensification of the India-Pakistani conflict is linked to other factors than COVID-19 (Staniland, 2020) . It is plausible that groups like the Taliban have attempted to exploit a situation where the state is weakened, border controls with Afghanistan get more difficult, dissatisfaction with the government's response is widespread (especially among religious groups), and a rise in poverty makes recruitment easier (Ahmed, 2020; Guz, 2020) . Concrete evidence of this is currently lacking, however. A reason for the slight bump in violent events in May and early June could be that many senior Taliban leaders and commanders became infect with SARS- CoV-2 (O'Donnell & Khan, 2020) . In the Philippines, the upwards trend in armed conflict events was mostly driven by a steep rise of clashes between the military and the Communist New People's Army (NPA). Both sides declared unilateral ceasefires when the number of infections increased rapidly in late March to facilitate responses to COVID-19. Accusing each other of continuing attacks, the government and the NPA decided not to extend their respective ceasefires in late April. There have been reports that the government utilized the distraction caused by the pandemic as well as the increased control gained during a strict lockdown for harsher measures against its opponents, including the rebels. Others claim that the NPA sought to utilize the pre-occupation of security forces with health-related tasks to launch further attacks and raid food supplies (Chavez, 2020; Lalu, 2020) . Besides its immediate health and economic effects, COVID-19 can also impact armed conflict risks, with these conflicts themselves being an important obstacle in dealing with the pandemic. This article provided an assessment of the impact of COVID-19 on armed conflict based on data from the first six months of 2020. Theoretically, the pandemic could affect conflict risks through increased grievances, possibilities to demonstrate solidarity, or modified opportunity structures for armed groups. Results show that in four of the nine countries under study, the number of armed conflict events declined after the onset of the COVID-19 crisis. These declines are mostly related to strategic decisions and less favourable opportunity structures for armed groups, such as logistical difficulties and attempts to increase popular support. They offer few prospects for health diplomacy and sustainable peacebuilding. In places like Afghanistan, where the Taliban restrained their military activities to gain local support, the initial decline might even set the stage for a later escalation of the armed conflict. Similar concerns exist regarding recruitment in Colombia and India. In five of the nine countries analysed, armed conflict prevalence increased in the face of the pandemic. This is further evidence that health diplomacy approaches demonstrating goodwill and reducing grievances have little impact during the pandemic (Polo, 2020) . COVID-19 did not change the root causes or principal dynamics of the armed conflicts in any of these five countries, but it accelerated existing trends and provided strategic opportunities for armed groups to exploit. Two factors are particularly relevant here: The weakening of state institutions (providing incentives for rebels to intensify military pressure) and a lack of (international) public attention (allowing to extend military operations without backlashes). While short-term rises in armed conflict risks related to the pandemic are mostly driven by changed opportunity structures, grievances could play a more prominent role when longer time horizons are considered. The economic repercussions associated with the current global spike in infections could exceed the coping capacities of households that did relatively well during the first COVID-19 wave. In coincidence with ethnic or religious cleavages, this could raise discontent to a level at which armed conflicts erupt. However, grievances usually take time to translate into organised armed activities. Declining levels of democracy as states claim emergency powers to combat COVID-19 are also a risk factor. Countries with a medium level of democracy and highly repressive regimes are empirically much more likely to experience civil wars than consolidated democracies (Cederman & Vogt, 2017) . Armed conflict can have tremendous negative effects on human security and health governance. It is therefore of crucial importance to monitor the impact of COVID-19 on armed conflict risks and to develop adequate policy responses, such as sanctioning armed groups trying to exploit the pandemic. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Impact of COVID-19 on conflict and peace in Pakistan How could coronavirus affect wars in the Arab world? Colombia's ELN rebels have 'no offensive plans' during coronavirus crisis COVID-19 Pandemic, a War to be Won: Understanding its Economic Implications for Africa. Applied Health Economics and Health Policy On malaria and the duration of civil war How Pakistan 'deep state' is using coronavirus cover to fuel terrorism in Kashmir Intelligence reports show Maoists are using lockdown time to regroup in Bastar COVID-19: a window of opportunity for negotiations with the ELN Dynamics and logics of civil war Predicting armed conflict: Time to adjust our expectations? Pathogens, weather shocks and civil conflicts Global Health Diplomacy as a Tool of Peace PRRD's anti-NPA tack a 'deflection tactic', says CPP COVID-19 threatening global peace and security, UN chief warns The bottom billion: Why the poorest countries are failing and what can be done about it Afghanistan's war on Covid-19: politics of the pandemic COVID-19 in Africa: Driver of conf lict, or too early to tell Ethnicity, insurgency, and civil war Libya's civil war has a rare shot at peaceful ending COVID-19 not affecting Afghan border fencing, Pakistan army says ISIS offensive exploits pandemic The impact of environmental cooperation on peacemaking: Definitions, mechanisms and empirical evidence World Economic Outlook Caught between virus and violence in Indian Kashmir For the Taliban, the pandemic is a ladder Covid-19 in India and Sri Lanka: new forms of Islamophobia The Taliban makes the most of Covid-19 crisis in Afghanistan Thailand: killings of insurgents ends southern separatist ceasefire Do health interventions support peace through ''disaster diplomacy COVID-19 and left wing extremism in India: emerging concerns HIV/AIDS, life expectancy, and the opportunity cost model of civil war Terrorism is top PH threat Do we really know that Chinese COVID-19 statistics are being manipulated? Armed groups in Colombia target children amid pandemic The effect of the Covid-19 pandemic on global armed conflict: Early evidence Yemen's rebels crack down as COVID-19 and rumors spread From bad to worse: The impact(s) of Covid-19 on conflict dynamics Leader of Afghan Taliban said to be gravely ill with the Coronavirus Thai rebels pay a price for coming above ground Organized violence A pandemic of violence? The impact of COVID-19 on conflict Contagion and chaos: Disease, ecology, and national security in the era of globalization Introducing ACLED: An armed conflict location and event dataset Anti-lockdown protests around the world Coronavirus Pandemic (COVID-19 Are the coronavirus ceasefires working? Oslo: PRIO Islamic State seeks comeback under cover of coronavirus Masters of their domains: The role of state capacity in civil wars Targeting environmental infrastructures, international law, and civilians in the new Middle Eastern wars Kashmir, India, and Pakistan and coronavirus Why do civil wars occur? Another look at the theoretical dichotomy of opportunity versus grievance Women and children living in areas of armed conflict in Africa: A geospatial analysis of mortality and orphanhood Civil war & the global threat of pandemics World Bank predicts sharpest decline of remittances in recent history Research for this article was supported by the Australian Research Council (grant DE190101268) and the Albert Shimmins Fund. I thank Immanuel Jacobi for his support with the data analysis as well as the editors and reviewers of World Development, Sol Santos and Arden Haar for helpful feedback on previous versions of this study.