When estimating the impact of conflict, we will apply the framework utilized by impacts and risk assessments. This framework defines risks broadly, as a function of hazard, exposure and vulnerability:

Risk = Hazard * Exposure * Vulnerability      (1)

This formulation is commonly used to estimate the impacts of climatic changes and has become the gold standard to assess populations’ and communities’ vulnerability to natural hazards and shocks [1, 2]. It is simple, flexible and generalizable, and enjoys wide consensus within the natural- science research community. An early application of this to the impact of conflict on humanitarian disasters is the  INFORM Severity Index, in part based on research work by Eriksson et al. [3, 4]. The programme will build upon and expand this work to provide an integrated, cross-sectoral, detailed evaluation of communities’ vulnerability to adverse humanitarian impacts.

Risk is also formulated as the probability of war (the hazard) and the consequences when it occurs [2]:

Risk = (probability of event) * (severity of consequences)        (2)

With hazard we mostly mean the probability of conflict of varying severity, although we will briefly also consider (in CC3) other hazards and treat conflicts as a source of vulnerability. We will refer to the consequence of conflict as observed as the impact.

Exposure refers to the inventory of elements in an area in which hazard events may occur and thus identifies the presence of people, livelihoods, infrastructure, economic, social, or cultural resources that could be adversely affected by the hazard [5]. Exposure is a key dimension of risk: humanitarian disasters occur only when hazards affect where people live.

Vulnerability indicates the propensity or predisposition to be adversely affected– a lack of re- sources and capacity to cope, adapt and recover from shocks which enhance adverse effects on the exposed communities. Generally, vulnerability refers to the potential for loss; as the magnitude of losses changes over time and space, and among different groups, vulnerability equally varies [6]. Vulnerabilities partly stem from social inequalities, which shape communities’ ability to respond and recover from shocks [7] and partly from place inequalities, including the level of urbanization, growth rates, and economic conditions [6].

Although often conflated, exposure and vulnerability are distinct. Exposure is a necessary, but not sufficient, condition for risk to materialize. Communities can be exposed but still avoid being vulnerable to hazards, at least partially, provided that they have means and tools to adapt and recover. For example, some can migrate out from the conflict area, others can not.

This risk framework enables us to provide a detailed assessment of humanitarian impacts of conflict in a broader sense than the one provided by only traditional formulations. For example, health status can be seen as a product of exposure to hazards, whereas vulnerability is a function of the financial and human resources and means available to address the risk-increasing conditions and the extent to which these resources are efficiently allocated [8].

We will link the estimate of conflict exposure to outcome-specific vulnerabilities. For instance, warfare itself exposes human populations through damages to power plants, water systems, and other infrastructure. Some socio-economic, gender, and age groups are likely to be disproportionately affected by these outcomes, as systemic discrimination increases their vulnerability to shocks [9]. Other exposures come indirectly. For instance, population displacements expose both migrant populations and the residents of the locations they move into to infectious diseases puts social cohesion under pressure and stretches infrastructures to provide water and health services to the limits.

For centuries, science has helped us understand the world around us by breaking systems into their components. Many important questions, however, can only be answered by looking at systems as units and observing the relationships across these units [10]. ‘Complex Systems’ is a scientific approach to study how relationships between units give rise to the collective patterns of behaviors of a system [11]. They are characterized by non-linearities, self-organization, and the ‘emergence’ of global properties and behaviors. These higher-order properties arise from the interactions among the system’s component, but cannot be understood when looking at each part in isolation [11].

According to this approach, social systems are complex adaptive systems, as they are open and able to adapt to their environment [12]. Due to interlinkages between components and to their openness and exchanges with the external environment, complex systems can be more vulnerable to adverse outcomes than simpler ones [13]. To gain a broad understanding of the impacts of armed conflict, we will adapt a complex systems approach, accounting for how six interlinked ‘systems’ interact with each other. These are public health, economics, water, social-psychological aspects, political institutions, and forced population movements. These effects of war are typically studied in isolation, but they clearly interact. To study the whole system in its complexity and understand the emergence of systemic patterns of behavior, the programme will be organized as a combination of ‘single-system’ work packages and a set of cross-cutting packages that bind these together.

The health system that is the focus of WP1, for instance, explores how individuals’ health responds to disease vectors and prevalence of diseases in their geographical proximity, but also to the provision of public health such as immunization programmes and maternal health care, nourishment, and access to water. As such, the health system critically depends on the economic system at the core of WP3 and WP4, as resources available for public health provision diminish due to destruction, diversion, dis-saving, and disruption [14]. Likewise, health is related to the water system studied in WP5 [15, 16]. Access to clean water decreases early childhood mortality and waterborne infectious diseases [17–19].

We will also study indirect effects. Impacts of conflict on water provision, for instance, are detrimental for agricultural economics and fuel internal displacement (R1). Outside of combat zones, the lack of fuel for groundwater pumping, damaged water pipes, or dysfunctional waste- water treatment poses serious consequences for local populations, often forcing them to flee [20]. The recent cholera outbreak in Yemen has been the worst such epidemic in modern history [21, 22] and further driven displacement. Armed conflict also diminishes vaccinations efforts and exacerbates the impact of Ebola [23, 24].

Armed conflict also affects psychological health and social cohesion, the topics of what we call the ‘social-psychological system’ (WP2). This system also interacts with the others. For instance, access to piped household water strengthens social integration and mental health [25]. Trust and social cohesion are important to economic growth and to functioning political systems.

By studying the systems jointly, we will be better able to understand and anticipate when conse- quences of violence are particularly disastrous. For actors that seek to mitigate these consequences, an analysis of how the systems interact may suggest what actions are particularly effective.

1. IPCC. Climate Change 2014: Impacts, Adaptation, and Vulnerability 2014.
2. IPCC (eds Field, C. et al.) (Cambridge University Press, 2012).
3. Eriksson, A., Ohlsén, Y. K., Garfield, R. & von Schreeb, J. Who Is Worst Off? Developing a Severity-scoring Model of Complex Emergency Affected Countries in Order to Ensure Needs Based Funding. eng. PLoS currents 7 (Nov. 2015).
4. Eriksson, A. et al. How Bad Is It? Usefulness of the “7eed Model” for Scoring Severity and Level of Need in Complex Emergencies. eng. PLoS currents 8 (June 2016).
5. Cardona, O. et al. Determinants of risk: exposure and vulnerability (Cambridge University Press, Cambridge, 2012).
6. Cutter, S. L., Boruff, B. J. & Shirley, W. L. Social Vulnerability to Environmental Hazards*. Social Science Quarterly 84, 242–261 (2003).
7. Adger, W. N. & Kelly, P. M. Social Vulnerability to Climate Change and the Architecture of Entitlements. Mitigation and Adaptation Strategies for Global Change 4, 253–266 (Sept. 1, 1999).
8. Ghobarah, H. A., Huth, P. K. & Russett, B. M. Civil wars kill and maim people–Long after the shooting stops. American Political Science Review 97, 189–202 (2003).
9. Llorente-Marrón, M., Díaz-Fernández, M., Méndez-Rodríguez, P. & González Arias, R. Social Vulnerability, Gender and Disasters. The Case of Haiti in 2010. Sustainability 12 (2020).
10. Bar-Yam, Y. Dynamics of complex systems OCLC: 246993529. 848 pp. (The Advanced Book Program, Addison-Wesley, Reading, Massachusetts, 1997).
11. Serra, R. & Zanarini, G. Complex Systems and Cognitive Processes OCLC: 1076231826 (Springer Berlin / Heidelberg, Berlin, Heidelberg, 2013).
12. Chan, S. Complex Adaptive Systems Research Seminar in Engineering Systems. 2001.
13. Gell-Mann, M. Simplicity and Complexity in the Description of Nature. Engineering and Science 51, 2–9 (1988).
14. Collier, P. On the economic consequences of civil war. Oxford Economic Papers – New Series 51, 168–183 (1999).
15. UNEP. Africa Water Atlas (Division of Early Warning and Assessment, United Nations Environment Programme (UNEP), Nairobi, 2010).
16. UN Water. The United Nations World Water Development Report 2019: Leaving No One Behind (UNESCO, Paris, France, 2019).
17. Merrick, T. W. The Effect of Piped Water on Early Childhood Mortality in Urban Brazil, 1970 to 1976. Demography 22, 1 (1985).
18. Duflo, E., Greenstone, M., Guiteras, R. & Clasen, T. Toilets Can Work: Short and Medium Run Health Impacts of Addressing Complementarities and Externalities in Water and Sanitation tech. rep. (NBER, 2015).
19. Fewtrell, L. et al. Water, sanitation, and hygiene interventions to reduce diarrhoea in less developed countries: a systematic review and meta-analysis. The Lancet Infectious Diseases 5, 42–52 (2005).
20. ICRC. Bled dry. How war in the Middle East is bringing the region’s water supplies to breaking point. An ICRC report. (International Committee of the Red Cross, 2015).
21. Camacho, A. et al. Cholera epidemic in Yemen, 2016–18: an analysis of surveillance data. The Lancet Global Health 6, e680–e690 (2018).
22. Ng, Q. X., Deyn, M. L. Z. Q. D., Loke, W. & Yeo, W. S. Yemen’s Cholera Epidemic Is a One Health Issue. Journal of Preventive Medicine and Public Health 53, 289–292 (2020).
23. Ngo, N. V. et al. Armed conflict, a neglected determinant of childhood vaccination: some children are left behind. Human Vaccines & Immunotherapeutics 16, 1454–1463 (2019).
24. Wells, C. R. et al. The exacerbation of Ebola outbreaks by conflict in the Democratic Republic of the Congo. PNAS 116, 24366–24372 (2019).
25. Devoto, F. et al. Happiness on Tap: Piped Water Adoption in Urban Morocco. American Economic Journal: Economic Policy 4, 68–99 (2012).

WP lead: Anneli Eriksson, Johan von Schreeb, Debarati Guha

This WP will focus on understanding the health impact of armed conflict by analyzing the impact pathways. To define strategies to mitigate the health impact of conflict, such an understanding of causes of death and poor health are necessary. To achieve this, we will quantify impacts on mortality, causes of deaths, and basic public-health services, using a variety of methodological approaches. The package will have several components:

• We will develop the methods, concepts, and proxy indicators for vulnerability and exposure developed by the team [1, 2], working in close connection with CC3. We will define a set of proxy indicators for health and well-being in conflict contexts and their availability and aggregation level in regular surveys (DHS and others), extract data and assess progress over time through time-series analysis. To set a baseline, we will compare with similar vulnerability contexts without ongoing armed conflict. We will link statistically a broad range of indicators for health outcomes such as mortality, morbidities, levels of acute and chronic malnutrition, vaccination coverage, and provision of health services (deliveries and C-sections) to conflict exposure and intensity as defined in CC2 for 20–25 contemporary armed conflicts. We will also look more closely at two contemporary conflicts that differ in terms of pre-war income levels. Here, data will be studied at a regional level and related to geographic proximity to ongoing armed conflict. We will focus on malnutrition as an important driver of child mortality and measles vaccination coverage, as a proxy for health service availability and accessibility [3]. This study will be complemented with a qualitative study seeking to understand more of the mechanisms behind the health impact, through focus group discussions and interviews with affected people. 
  
  
• Verbal Autopsy and cause of death (CoD). Death registration data, particularly in com- plex emergencies, are notoriously sparse. Most deaths occur outside hospitals and remain unknown and unreported. Checchi et al. [4] have recently highlighted the absence of vali- dated methods as a main underlying reason for the lack of data on causes of death (CoD). To understand these patterns better in humanitarian settings, a variation of the verbal autopsy approach (VA) is a promising way to estimate CoD in populations where most deaths are undocumented. The quality and effectiveness of humanitarian response can only be improved by a more refined understanding of the underlying CoD, both in children and in adults. This information will also elucidate key inequalities related to other factors, such as age, gender and social status. We will undertake the testing experiment in Cox’s Bazar in Bangladesh. 
  
  
• Estimating impact on key impact indicators by means of Bayesian models. Estimating conflict impact on mortality and other indicators is necessary but widely acknowledged as a challenging task. In many instances, estimations remain non-quantified and are often us- ing broad qualitative descriptors such as ‘tens of thousands’ deaths. We will work to improve estimates for Yemen, DR Congo and Ethiopia through Bayesian modeling, lever- aging the ability of this approach to combine data from several sources that are heterogeneous in their original structure. By pooling several sources of evidence, some of them with too few observations to stand well on their own, we will be able to considerably improve estimates of excess mortality and other indicators in these three countries. We will also estimate biases inherent in the presence of different data sources and use imputation techniques to fill in for missing data [5].

WP lead: Jonathan Hall

This WP aims to assess the impact of exposure to violence on cooperative behavior in the form of political and social trust, political participation, attitudes towards political and economic institutions, and subjective well-being (SWB). We will rely upon two complementary sources of data. The first regards pro-social behavior in experimental games. We will provide the first truly global meta-analysis of the experimental games literature, incorporating data from conflict-affected societies, largely neglected in such analysis [cf. 6]. It will expand on the study of Bauer et al. [7] by incorporating new studies with clear in-group/outgroup treatments mapped onto war cleavages and measures of exposure to violence on the individual level. In combination with UCDP conflict data, we can examine the effects of war exposure on cooperative behavior in experimental games using multi-level modeling analytical strategies. The WP will make use of demographic characteristics such as gender included in surveys as well as experimental treatments designed to capture the in-group/outgroup dynamics of cooperative behavior. The second data source is the Life in Transition Survey [LiTS 8] that has recorded geographic location, socio-economic status, perceptions on social, economic and political issues for thousands of households and individuals across about 30 countries over 10 years. The dataset allows linking a rich set of measures of social cooperation at both the local and national level to several measures of conflict exposure. The work on impacts on SWB will be linked to other WPs through its importance for creativity, longevity and productivity [9] and political participation and voter turnout [10–12].

WP lead: Tilman Brück

This WP will review and estimate the micro-economic impacts of conflict across sectoral or topical domains. We will proceed in three steps. First, we will conduct a meta-analysis of the maturing literature on the micro-level impacts of violent conflict on people and households, updating and expanding Blattman & Miguel [13] and Verwimp et al. [14]. Second, as this literature mostly disregards impacts across domains, we hypothesize that the sum of all known within-domain effects on say growth is lower than the sum of all effects (including the impacts and interactions across domains). We will thus postulate and estimate a micro-macro model of cross-sectoral impacts of armed violence, providing a novel micro-founded estimate of the aggregate costs of conflict. Third, we will feed the findings of this model into the extended ViEWS model (CC4) to predict the near-term human development impacts of armed violence across sectors.

WP lead: Hannes Mueller

Economic development and poverty are adversely affected both during and after armed con- flict, through the humanitarian crisis triggered and the contraction of investments due to instability. Risk perceptions play a key role here, as they spread the costs of armed conflict beyond the conflict period [15]. If investments are affected by uncertainty and uncertainty increases before the conflict, they may decline even before violence breaks out. Conflicts will also make countries unable to attract investments after the conflict has ended because conflict risks stay high [16, 17]. In this WP, we will provide estimates of economic policy uncertainty at the monthly level for more than 180 countries to gauge the effect of conflict risk on investment incentives. Our methodology will build on two pillars. The first one is the literature on the economic cost of economic uncertainty [18]. Such uncertainty is another impact of armed conflict, as shown for the Spanish civil war case in García-Uribe et al. [19]. We will build on the seminal work of Baker et al. [20] who derive a monthly economic policy uncertainty (EPU) index for 22 countries. Their methodology is a dictionary-based method that counts economic, policy, and uncertainty terms in local newspaper sources and combines them in an index of uncertainty. We will reconstruct their index using a corpus of over 4 million newspaper articles covering 150 countries, training a machine-learning model on the 22 countries in Baker et al. [20] and predicting EPU across all the remaining countries. The results will allow us for the first time to track economic policy uncertainty in countries that are affected by armed conflict. We will validate our results by looking at sudden increases of the risk which were not followed by an actual outbreak of violence, i.e. the false positives in the conflict forecast model data. Next, we will then rely on the well-established literature on the effects of EPU on investments to gauge the costs of conflict in terms of foregone investment due to conflict risk.

WP lead: Ashok Swain

This WP will study how access to safe drinking water is affected by armed conflict, either due to destruction and contamination, or because insecurity hinders populations to reach the best sources. Among the mechanisms explored will be how disruptions to power supplies affect water storage and delivery systems, groundwater withdrawal or purification plants that depend on such supply [21, 22]. The WP will also study how large-scale forced population displacements lead to short-term (municipal- and industrial supply) and long-term (agricultural) changes in demand for water not only in conflict-affected areas but also in regions hosting these migrants. Post-conflict reconstruction efforts usually support and promote building large water infrastructure projects to stimulate food and energy production and economic recovery [23, 24]. The WP will investigate how water bodies, water transport systems, treatment plants, dams, and irrigation facilities are either intentionally or as collateral damage being destroyed or damaged by warring groups. For instance, in Iraq, ISIS used water as a weapon by withholding access, by flooding, and by contaminating water supplies [25, 26].

WP lead: Staffan Lindberg

Institutions that regulate access to power positions and how decisions are made and implemented are of crucial importance to many processes in the programme. This WP will contribute to several of these. First, it will study the impact of armed conflict on various aspects of political institutions. Using indicators from the Varieties of Democracy project [27], it will distinguish between the effects on elections, on institutions ensuring legislative or judicial constraints, and on civil society and freedom of speech and association. As a second goal, the WP will also explore how these institutions work to help preventing armed conflict in the first place [28-30], to establish a sound counterfactual for the impact of armed conflict, and to feed in to the conflict forecasting in CC4. Finally, the WP will work with CC3 to model how political institutions affect communities’ vulnerability to the impact of armed conflict.

1. Eriksson, A., Ohlsén, Y. K., Garfield, R. & von Schreeb, J. Who Is Worst Off? Developing a Severity-scoring Model of Complex Emergency Affected Countries in Order to Ensure Needs Based Funding. eng. PLoS currents 7 (Nov. 2015).
2. Eriksson, A. et al. How Bad Is It? Usefulness of the “7eed Model” for Scoring Severity and Level of Need in Complex Emergencies. eng. PLoS currents 8 (June 2016).
3. Checchi F; Gayer, M. G. R. M.-E. J. Public health in crisis-affected populations. A practical guide for decision-makers. Humanitarian Practice Network (61 2007).
4. Checchi, F. et al. Public health information in crisis-affected populations: a review of methods and their use for advocacy and action. The Lancet 390. Publisher: Elsevier, 2297–2313 (Nov. 18, 2017).
5. Manzi, G. et al. Modelling bias in combining small area prevalence estimates from multiple surveys. Journal of the Royal Statistical Society: Series A (Statistics in Society) 174, 31–50 (2011).
6. Balliet, D., Wu, J. & De Dreu, C. K. W. Ingroup favoritism in cooperation: A meta-analysis. Psychological bulletin 140, 1556–1581 (6 2014).
7. Bauer, M. et al. Can War Foster Cooperation? Journal of Economic Perspectives 30, 249–74 (2016).
8. European Bank for Reconstruction and Development. Life in transition: A decade of measuring transition 2017.
9. De Neve, J.-E., Diener, E., Tay, L. & Xuereb, C. The Objective Benefits of Subjective Well- Being (August 6, 2013). in World Happiness Report 2013 (eds Helliwell, J., Layard, R. & Sachs, J.) (New York: UN Sustainable Development Solutions Network, 2013).
10. Dolan, P., Metcalfe, R. & Powdthavee, N. Electing happiness: does happiness affect voting and do elections affect happiness Discus. Pap. Econ. 30. 2008.
11. Liberini, F., Redoano, M. & Proto, E. Happy voters. Journal of Public Economics 146, 41–57 (2017).
12. Ward, G. Is happiness a predictor of election results? CEP Discussion Papers, CEPDP1343. Centre for Economic Performance, London School of Economics and Political Science, London, UK. 2015.
13. Blattman, C. & Miguel, E. Civil War. Jounal of Economic Literature 48, 3–57 (2010).
14. Verwimp, P., Justino, P. & Brück, T. The microeconomics of violent conflict. Journal of Development Economics 141, 102297 (Nov. 2019).
15. Besley, T. & Mueller, H. Estimating the Peace Dividend: The Impact of Violence on House Prices in Northern Ireland. American Economic Review 102, 810–33 (2012).
16. Mueller, H. How Much Is Prevention Worth? Background paper for United Nations–World Bank Flagship Study, Pathways for Peace: Inclusive Approaches to Preventing Violent Conflict, World Bank, Washington, DC. 2017.
17. Rohner, D. & Thoenig, M. The Elusive Peace Dividend of Development Policy: From War Traps to Macro-Complementarities. Annual Review of Economics (2020).
18. Bloom, N. The Impact of Uncertainty Shocks. Econometrica 77, 623–685 (2009).
19. García-Uribe, S., Mueller, H. & Sanz, C. Economic Uncertainty and Divisive Politics: Evidence from the dos Españas CEPR Discussion Paper: DP15479. 2020.
20. Baker, S. R., Bloom, N. & Davis, S. J. Measuring Economic Policy Uncertainty*. The Quar- terly Journal of Economics 131, 1593–1636 (July 2016).
21. Schillinger, J., Özerol, G., Güven-Griemert, & Heldeweg, M. Water in war: Understanding the impacts of armed conflict on water resources and their management. WIREs Water 7, e1480 (2020).
22. Grech-Madin, C. The Water Taboo: Restraining the Weaponization of Water in International Conflict. Uppsala: Department of Peace and Conflict Research. 2020.
23. Swain, A. Water and post-conflict peacebuilding. Hydrological Sciences Journal 61, 1313– 1322 (2016).
24. Döring, S. From Bullets to Boreholes: A Disaggregated Analysis of Domestic Water Cooperation in Drought-prone Regions. Global Environmental Change 65, 102147 (2020).
25. von Lossow, T. Water as Weapon: IS on the Euphrates and Tigris (Berlin: German Institute for International and Security Affairs., 2016).
26. Müller, M. F. et al. Impact of the Syrian refugee crisis on land use and transboundary freshwater resources. PNAS 113, 14932–14937 (2016).
27. Coppedge, M. et al. V-Dem Codebook v10 Varieties of Democracy (V-Dem) Project. 2020.
28. Muller, E. N. & Weede, E. Cross-national variations in political violence: A rational action approach. Journal of Conflict Resolution 34, 624–651 (1990).
29. Hegre, H., Ellingsen, T., Gates, S. & Gleditsch, N. P. Toward a democratic civil peace? Democracy, political change, and civil war, 1816–1992. American Political Science Review 95, 33–48 (2001).
30. Cederman, L.-E., Hug, S. & Krebs, L. F. Democratization and civil war: Empirical evidence. Journal of Peace Research 47, 377–394 (2010).

CC lead: Håvard Hegre This CC will coordinate the impact assessments in all the outcome-specific WPs, ensuring consistency in terms of methods, data, and definitions of outcome metrics and units of analysis, and binding the work together utilizing the risk framework. To the extent possible, the CC will cast outcomes from the WPs as estimates of the causal effect of conflict of a given magnitude on the outcome. We will take into account how outcomes relate to each other and depend on each other. The conflict impact will be measured as a probability distribution over deviations from levels that are normal to the units of analysis they refer to, looking to the work in CC3 that will also estimate these counter-factual trends. The output of the estimates in this CC will be fed into the early-warning system developed in CC4.

CC lead: Magnus Öberg This CC will work with the WPs and CC1 to identify the distribution of armed conflict impacts across social groups, age groups, and gender. To accurately model how many people are affected by violence we combine geographical information on population densities [1] with geo-coded events data measuring the intensity and location of different types of violence [2]. Using geographic information systems (GIS) we generate measures of how many people are affected by what type of violence and at what intensity in a given location or country for a given period of time. We then combine this with specific variables from CC3 that describe populations’ vulnerability to the impact of violence on different outcome variables (e.g. economic conditions, health conditions, age and gender structure, education levels, water access etc). We will test and evaluate various estimates of distance. Presumably, different types of violence and events with different intensities have differently sized impacts on the different outcome variables (e.g. on people’s health, economy, decisions to flee, etc) depending on the characteristics of the affected population and over time. The models of exposure will feed into the work of all the CCs.

CC lead: Paola Vesco The most severe humanitarian crises globally are found in places exposed to a combination of human and natural induced hazards [3], such as severe food crises in conflict-ridden South Sudan and Northern Nigeria. The combination of social and natural events can give rise to a cascade of temporally or spatially dependent risks whose consequences cannot be predicted by observing each of these events separately. However, the effects of these compound events tend to be studied in isolation, thus potentially underestimating the risks [4]. Likewise, knowledge of effective interventions to break the destructive interactions between natural and conflict-related hazards is limited [5]. This CC will adopt the innovative risk assessment method to estimate the risks of ‘compound events’ [4]. We will observe the processes that cause adverse humanitarian impacts as inter-related and interdependent, building a bridge between social and climate science to provide a better understanding of these complex events. The CC will estimate the underlying vulnerability component of our risk model, in the form of the statistical modeling of interactive impact, and as a composite indicator of vulnerability, to first assess conflict-induced vulnerability to adverse events. We will use the UCDP-GED conflict data, real-time surface weather data to construct agro-climatic indicators (ECWMF), novel data on the geographic location of climate-related disasters based on the widely-used EM-DAT database [6], food security indicators to measure the severity of disaster impact [7], as well as meteorological data giving exogenous indicators of disaster severity [cf. 8]. We will consider immediate effects as well as the legacy of terminated conflict activities.

CC lead: Håvard Hegre This CC will set up an early-warning system for conflict-driven humanitarian disasters to direct attention to ongoing and impending humanitarian disasters. Simulations of various conflict scenarios through the early-warning system are useful to display their effects and the likely benefits of their prevention. Finally, we will use the system as a focal point for the work in the various work packages, helping to coordinate the research and ensure that they feed into a common framework. The system will provide assessments of the risk of high levels of conflict-driven impact over the next three years, for the selected outcomes at the country level as well as for a 0.5×0.5 decimal degrees geographical grid as defined by  PRIO-GRID. The system will build on the infrastructure and procedures in  ViEWS, developed since 2017 with funding from the ERC [9]. The existing ViEWS system provides an excellent point of departure for the hazard component of the impact assessment. The programme will expand ViEWS by setting it up to also produce forecasts of the number of fatalities from conflict as defined by the UCDP. The main part of the armed conflict impact forecasts will combine the ViEWS forecasts with the estimates of impacts given realized conflict developed in CC1, applying the exposure model in CC2 and the underlying vulnerabilities estimated in CC3 to produce probabilistic estimates of humanitarian disasters/‘conflict impact’.

CC lead: Hannes Mueller This WP will provide global and country- specific estimates of the costs of conflict to promote preventive action. It will use the impacts of conflict developed in the outcome-specific WPs as a basis for a dynamic costing model. These will be fed into the methodology developed in Mueller [10] conducted for World Bank Group & United Nations [11], and gradually adapted to the forecasting model developed in CC4. The long-term impact of conflict outbreaks may thereby be estimated not only from the effects of contemporaneous violence levels but also from the possible escalations and future outbreaks that follow. Such a complete, forward-looking approach that takes into account future costs will allow us to calculate what the optimal intervention regime is. Prevention could be costly because it means expenditure before conflict breaks out, but it has strong dynamic benefits if it prevents countries from falling into the conflict trap [12].

1. Center for International Earth Science Information Network. Documentation for the Gridded Population of the World, Version 4 (GPWv4) Palisades NY: NASA Socioeconomic Data and Applications Center (SEDAC). 2016.
2. Sundberg, R. & Melander, E. Introducing the UCDP Georeferenced Event Dataset. Journal of Peace Research 50, 523–532 (2013).
3. FAO, IFAD, UNICEF, WFP, and WHO. The State of Food Insecurity Report 2017: Buidling Resilience for Peace and Food Security (2017).
4. Zscheischler, J. et al. Future climate risk from compound events. Nature Climate Change 8, 469–477 (2018).
5. Buhaug, H. & Uexkull, N. V. Vicious Circles: Violence, Vulnerability, and Climate Change.Annual Review of Environment and Resources forthcoming (2021).
6. Rosvold, E. & Buhaug, H. Geocoded Disasters (GDIS) Dataset, 1960-2018 (Preliminary Re- lease) Palisades, NY: NASA Socioeconomic Data and Applications Center (SEDAC). 2020.
7. Cooper, M. W. et al. Mapping the effects of drought on child stunting. Proceedings of the National Academy of Sciences 116, 17219–17224 (2019).
8. Dellmuth, L. M. et al. Humanitarian Need Drives Multilateral Disaster Aid. Proceedings of the National Academy of Sciences of the United States of America (PNAS) forthcoming (2021).
9. Hegre, H. et al. ViEWS: A political Violence Early Warning System. Journal of Peace Re- search 56, 155–174 (2019).
10. Mueller, H. How Much Is Prevention Worth? Background paper for United Nations–World Bank Flagship Study, Pathways for Peace: Inclusive Approaches to Preventing Violent Conflict, World Bank, Washington, DC. 2017. 
11. World Bank Group & United Nations. Pathways for Peace: Inclusive Approaches to Preventing Violent Conflict. Main Messages and Emerging Policy Directions. (International Bank for Reconstruction and Development/The World Bank, 2017).
12. Collier, P. et al. Breaking the Conflict Trap. Civil War and Development Policy (Oxford University Press, Oxford, 2003).

The joint impact assessment in CC1 and the risk assessment in CC4 will be formulated at two levels of analysis: Countries as defined by Gleditsch & Ward [1] and Weidmann  et al. [2], and the PRIO-GRID cell according to Tollefsen et al.  [3]. For these, the aim throughout the programme will be to provide estimates of impact across all units globally, using techniques to handle missing data and to use estimates developed in the WPs based on samples as points of departure for estimating effects across entire conflict zones. In the outcome-specific work packages,we will adapt units of analysis according to data availability. For instance, we will link surveys, which typically do not have universal coverage, to geographic point coordinates or administrative regions.

In several parts of the programme, we will seek to quantify the universal, average (although conditional) impact of conflict on the various outcomes. This requires that we define a metric to compare the intensity of the violence that is causing the impacts, with uniform coverage and strict adhesions to definitions that do not vary over time or space. The leading source for such data is the Uppsala Conflict Data programme [4], that have collected the number of recorded fatalities in all armed conflicts since 1989 that caused at least 25 battle-related deaths. The UCDP-GED dataset [5], that records all events within these conflicts with precise indications of where and when vio- lence occurred as well as estimates of how many were killed in direct violence. The data-collection component will collaborate with R2 and R3. To measure the outcomes, we will make use of data from the World Bank Indicators, from the Global Burden of Disease project, the WHO, and survey data from the CE-DAT [6] and EM-DAT databases [7] as well as from LSMS [7], DHS [7], MICS [10], and LiTS [11]. Where feasible, we will use metrics and sources that allow analyzing impact separately by gender and by social group. In survey data, such characteristics are often included in aggregates or individual records, or can be inferred from the geographical location of the respondents.

The programme will systematically handle missing and incomplete data. To compensate for uncertainty about the exact location of some conflict events, we will relate the impact to the probability distribution over conflict locations. This is computed using multiple imputation [12]. Other variables used in the programme, either as outcomes or as core conditional variables, also contain missing observations. This is particularly the case for survey data [13]. To avoid biased parameter estimates and/or standard errors [14], we will use multiple-imputation techniques (e.g.the Amelia II package).

The programme will summarize the existing empirical research on the impact of conflict through a meta-analysis – the statistical analysis of published research findings on a given hypothesis or empirical effect [15, 16–18]. Following these studies, we will i) define systematic criteria for including/excluding the studies and estimated effects; ii) do a broad systematic search of the literature to identify all studies that meet the criteria; iii) collect textual information and its conversion to quantitative features characterizing the studies; iv) analyze statistically the empirical findings in the selected sample.

Earlier studies [e.g. 19] typically estimate impact by in effect limiting exposure to the area within 50km of UCDP events. To provide a better model, both as input to modeling in other parts of the programme and as a deliverable in itself, we will work out a set of conflict exposure models. As a first step, the impact zone will be modeled as circles around the conflict events and functions that let impact decay over time, adding the size of local populations. We will then explore several extensions to such models – parameterize them and use statistically estimate precisely how the impact of conflict is reduced with a longer distance to conflict events in space and time. To identify the effect, we will use the observed outcomes treated in the WPs described below, along with a model for expected outcomes in a location in the absence of conflict. We will also consider more advanced concepts of distance, taking for instance into account that the impact of conflict may fail to cross international borders and be more likely to move through areas with contiguous settlements than across sparsely populated regions. We will explore network models or more complicated models of linkages between locations, and validate using out-of-sample evaluation of predictive performance. Finally, taking the probabilistic hazard estimates derived in the early-warning model as estimates, we will explore the importance of latent (unobserved) armed conflict.

CC3 sets out to construct a composite Vulnerability Index (VI) [20] at both the national and the sub-national level. Adapting the model developed by Eriksson et al. [21], the VI will be math- ematically derived by aggregating a combination of variables that represent different dimensions of communities’ vulnerability [22]. The main sub-components will be institutional, social and economic vulnerability/resilience, and community capital [23]. First, we will identify the relevant and robust variables to represent each sub-dimension of vul- nerability using the data collected in the programme. Next, the variable scores in each sub-index will be averaged to obtain an individual score for each sub-dimension of vulnerability (e.g. eco- nomic), and further aggregated to get the final VI [22, 24]. We will perform extensive sensitivity and robustness tests to evaluate how methodological choices influence the scores [25], including validating against existing indices such as the Social Vulnerability Index [26] and the Predictive Indicator of Vulnerability [27]. The VI will be used together with the early-warning system (CC4) to provide impact assessments for conflict and non-conflict scenarios. Hence, the VI will serve as both an ex-ante counterfactual and as an assessment of the conflict impact on societal and community vulnerability. As a final step, the Vulnerability assessment will be used to assess the risk of natural disasters and climatic shocks in conflict-affected countries according to IPCC’s definition. To compute risk, we will integrate the conflict-scenario vulnerability assessment, with the esti- mates of hazard and exposure. We will collect data on the severity of climatic shocks and natural disasters to give an accurate representation of hazards, whereas demographic data including inhab- ited area, population size and distribution at both national and individual level will be included to proxy exposure. This will enable us to provide a thorough assessment of compound risks [28], induced by the combination of conflict exposure and natural disasters.

We will use Bayesian models to estimate malnutrition, vaccination coverage, and excess mortality. These are useful in the sparse-data situation in conflict settings, as they allow effectively accounting for all sources of variability, incorporate past knowledge from previous studies (‘priors’), handling of missing data, and generate an easily interpretable direct posterior probabilistic value [29].

To estimate causes of death (CoD) in conflict settings, the programme will adapt verbal autopsy (VA) approaches to conflict settings. These build on face-to-face interviews with a close relative of the deceased, collecting information on signs, symptoms, and circumstances leading to death, which is then interpreted into CoD. The increasing use of VA to collect routine CoD data in stable, low- income settings has led to many versions of the instrument such as the WHO VA 2016. While these tools are commonly used in low- and middle-income countries, their use in humanitarian settings with severe challenges in terms of security and access is novel and therefore, must be simplified, adapted and tested. We will test in a field experiment a reformulated VA tool that can: i) be used in complex emergency settings and ii) add and validate data to survey tools in conflict settings. We will compare VA results in a refugee camp with hospital records to validate an adapted instrument.

With incomplete input data and heavy reliance on various estimation techniques, it is necessary to account for the uncertainty in the results. This is often ignored in the literature. The study of Cooper  et al. [30, SI, p. 8], for instance, ignores uncertainty completely. We will make use of a suite of techniques to cover various sources of uncertainty. Where available standard pack- ages do not provide such estimates, we will use bootstrapping techniques [31] to assess statistical uncertainty. We will make use of model ensembles to decrease the uncertainty about model specifi- cations, and apply simulation approaches from the early-warning system to bind the various sources of uncertainty together.

To complete our conflict risk assessment, we will combine the estimates of impacts of observed conflict, of exposure, and vulnerability with modeling of the probability of conflicts of varying intensity. This is necessary for two reasons. First, according to the ‘early warning, early action’ doctrine prevention is better than mitigation [32]. Moreover, even when prevention is impossible, an estimate of the likely future scale and duration of the conflict is important. The risk assessment has to be forward-looking and take uncertainty into account. To do this, we will extend the  ViEWS system [33, 34] to have global coverage, to forecast the number of fatalities in addition to the dichotomous presence/non-presence of conflict, and leverage the news-reading component described in WP4 to strengthen the warning of new conflicts [35]. The estimated probability distributions over conflict severity will be combined with the estimates of impact, exposure, and vulnerability to provide maps over the range of likely humanitarian effects of conflict. The early-warning system will be maintained as a live system that produces a monthly as- sessment of the risk of humanitarian disasters. The system can also be used to develop scenario assessments (CC5).

1. Gleditsch, K. S. & Ward, M. D. A Revised List of Independent States since the Congress of Vienna. International Interactions 25, 393–413 (1999).
2. Weidmann, N. B., Dorussen, H. & Gleditsch, K. S. The Geography of the International System: The CShapes Dataset. International Interactions 36, 86–106 (2010).
3. Tollefsen, A. F., Strand, H. & Buhaug, H. PRIO-GRID: A unified spatial data structure. Journal of Peace Research 49, 363–374 (2012).
4. Pettersson, T. & Öberg, M. Organized violence, 1989–2019. Journal of Peace Research 57, 597–613 (2020).
5. Sundberg, R. & Melander, E. Introducing the UCDP Georeferenced Event Dataset. Journal of Peace Research 50, 523–532 (2013).
6. Altare, C. & Guha-Sapir, D. The Complex Emergency Database: A Global Repository of Small-Scale Surveys on Nutrition, Health and Mortality. PLoS ONE 9 (ed Welte, A.) e109022 (Oct. 21, 2014).
7. Université catholique de Louvain (UCL) – CRED, D. Guha-Sapir. EM-DAT: The Emergency Events Database Brussels, Belgium,  www.emdat.be. 2020.
8. The World Bank, Center for Data Development. Living Standards Measurement Study Wash- ington DC,  https://www.worldbank.org/en/programs/lsms. 2020.
9. The DHS Program website. Demographic and Health Surveys Funded by USAID,  url://www.dhsprogram.com. 2018.
10. UNICEF. Multiple Indicator Cluster Surveys New York, USA, https://mics.unicef.org. 2021.
11. European Bank for Reconstruction and Development. Life in Transition Survey (LITS) Lon- don, UK, https://www.ebrd.com/what-we-do/economic-research-and-data/data/lits.html. 2016.
12. Croicu, M. & Hegre, H. A Fast Spatial Multiple Imputation Procedure for Imprecise Armed Conflict Events Typescript, Uppsala University/ViEWS.  http://www.pcr.uu.se/research/ views. 2018.
13. Manzi, G. et al. Modelling bias in combining small area prevalence estimates from multiple surveys. Journal of the Royal Statistical Society: Series A (Statistics in Society) 174, 31–50 (2011).
14. Allison, P. D. Missing Data in The SAGE handbook of quantitative methods in psychology (eds Millsap, R. E. & Maydeu-Olivares, A.) (Sage Publications, 2009).
15. Bauer, M. et al. Can War Foster Cooperation? Journal of Economic Perspectives 30, 249–74 (2016).
16. Stanley, T. D. & Doucouliagos, H. Meta-regression analysis in economics and business 186 pp. (Routledge, New York, 2012).
17. Stanley, T. D. & Doucouliagos, H. Neither fixed nor random: weighted least squares meta- regression: Weighted Least Squares Meta-Regression Analysis. Research Synthesis Methods 8, 19–42 (Mar. 2017).
18. Vesco, P., Dasgupta, S., Cian, E. D. & Carraro, C. Natural resources and conflict: A meta- analysis of the empirical literature. Ecological Economics 172, 106633 (2020).
19. Wagner, Z. et al. Armed conflict and child mortality in Africa: a geospatial analysis. The Lancet 392, 857 –865 (2018).
20. Diffenbaugh, N. S., Giorgi, F., Raymond, L. & Bi, X. Indicators of 21st century socioclimatic exposure. Proceedings of the National Academy of Sciences 104, 20195–20198 (Dec. 18, 2007).
21. Eriksson, A., Ohlsén, Y. K., Garfield, R. & von Schreeb, J. Who Is Worst Off? Developing a Severity-scoring Model of Complex Emergency Affected Countries in Order to Ensure Needs Based Funding. eng. PLoS currents 7, ecurrents.dis.8e7fb95c7df19c5a9ba56584d6aa2c59 (Nov. 2015).
22. Handbook on constructing composite indicators: methodology and user guide (ed OECD) OCLC: ocn244969711 (OECD, Paris, 2008). 158 pp.
23. Cutter, S. L., Burton, C. G. & Emrich, C. T. Disaster Resilience Indicators for Benchmarking Baseline Conditions. Journal of Homeland Security and Emergency Management 7 (2010).
24. Chen, T.-Y. An outcome-oriented approach to multicriteria decision analysis with intuition- istic fuzzy optimistic/pessimistic operators. Expert Systems with Applications 37, 7762 –7774 (2010).
25. Marzi, S. et al. Constructing a comprehensive disaster resilience index: The case of Italy.PLOS ONE 14 (ed Linkov, I.) e0221585 (Sept. 16, 2019).
26. Cutter, S. L., Boruff, B. J. & Shirley, W. L. Social Vulnerability to Environmental Hazards*.Social Science Quarterly 84, 242–261 (2003).
27. Adger, W. et al. New Indicatorsof Vulnerability and Adaptive Capacity Tyndall Centre for Climate Change Research. Technical Report. Norwich, UK: 2004.
28. Zscheischler, J. et al. Future climate risk from compound events. Nature Climate Change 8, 469–477 (2018).
29. Dunson, D. B. Commentary: Practical Advantages of Bayesian Analysis of Epidemiologic Data. American Journal of Epidemiology 153, 1222–1226 (June 2001).
30. Cooper, M. W. et al. Mapping the effects of drought on child stunting. Proceedings of the National Academy of Sciences 116, 17219–17224 (2019).
31. Efron, B. & Tibshirani, R. The Bootstrap Method for Assessing Statistical Accuracy. Behaviormetrika 12, 1–35 (1985).
32. World Bank Group & United Nations. Pathways for Peace: Inclusive Approaches to Preventing Violent Conflict. Main Messages and Emerging Policy Directions (International Bank for Reconstruction and Development/The World Bank, 2017).
33. Hegre, H. et al. ViEWS: A political Violence Early Warning System. Journal of Peace Re- search 56, 155–174 (2019).
34. Hegre, H. et al. ViEWS2020: Revising and evaluating the ViEWS political Violence Early- Warning System. Journal of Peace Research In press (2021).
35. Mueller, H. & Rauh, C. Reading Between the Lines: Prediction of Political Violence Using Newspaper Text. American Political Science Review 112, 358–375 (2018).