Terrorism and child mortality

Meierrieks, Daniel; Schaub, Max Article — Published Version Terrorism and child mortality Health Economics Provided in Cooperation with: John Wiley & Sons Suggested Citation: Meierrieks, Daniel; Schaub, Max (2024) : Terrorism and child mortality, Health Economics, ISSN 1099-1050, Wiley, Hoboken, NJ, Vol. 33, Iss. 1, pp. 21-40, https://doi.org/10.1002/hec.4757 This Version is available at: https://hdl.handle.net/10419/288238 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by/4.0/ wileyonlinelibrary.com/journal/hecHealth Economics. 2024;33:21–40. 21 1 | INTRODUCTION Terrorism has become more common and deadly after 2001. 1 This is especially true for Africa, a continent that, over the last 15years, has seen a spike in terrorist activity (Gaibulloev & Sandler,2019). At the same time, the continent is also plagued by comparatively high and persisting levels of child mortality (Burstein etal.,2019). Recognizing this concurrence, we ask whether terrorism leads to an increase in child mortality, and if so, why. To this effect, we study the impact of terrorism on child mortality for a sample of 52 African countries between 2000 and 2017. We combine geo-coded data on terrorism with highly spatially disaggregated data on child mortality and morbidity at the 0.5×0.5° grid level. This allows us to track child mortality and its likely causes on a sub-national level. We contribute to the literature in several ways. First, by using high-resolution data for much of Sub-Saharan Africa and parts of Northern Africa over a long time span, we provide systematic evidence on the terrorism-child mortality nexus in this part of the world. Thus, our study complements a small number of case-studies on the consequences of terrorism for child health in Africa. For Burkina Faso and Nigeria, this evidence indicates that terrorism impedes access to perinatal healthcare (Chukwuma & Ekhator-Mobayode,2019; Druetz etal.,2020). For Cameroon, it has been shown that the Boko Haram insurgency is associated with lower height-to-weight ratios in children under five, likely caused by infectious diseases and the underutilization of health services (Kaila etal.,2021). We also add to extant research on the potentially adverse effects of terrorism on child health more broadly. For example, several empirical studies find that in utero exposure to terrorist attacks is associated with lower infant weight due to maternal stress and poor nutrition (e.g., Camacho,2008; Lauderdale,2006; Mansour & Rees,2012; 1WZB Berlin Social Science Center, Berlin, Germany 2University of Hamburg, Hamburg, Germany Correspondence Max Schaub. Email: [email protected] Abstract How does terrorism affect child mortality? We use geo-coded data on terrorism and spatially disaggregated data on child mortality to study the relationship between both variables for 52 African countries between 2000 and 2017 at the 0.5×0.5° grid level. Our estimates suggest that moderate increases in terrorism are linked to several thousand additional annual deaths of children under the age of five. A panel event-study points to economic effects that are larger and compound over time. Interrogating our data, we show that the direct impact of terrorism tends to be very small. Instead, we theorize that terrorism causes child mortality primarily by triggering adverse behavioral responses by parents, medical workers, and policymakers. We provide tentative evidence in support of this argument. KEYWORDS Africa, child mortality, panel event-study, terrorism JEL CLASSIFICATION D74, I10, I12 RESEARCH ARTICLE Terrorism and child mortality Daniel Meierrieks1 | Max Schaub1,2 DOI: 10.1002/hec.4757 Received: 6 November 2022 Revised: 25 August 2023 Accepted: 29 August 2023 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2023 The Authors. Health Economics published by John Wiley & Sons Ltd. MEIERRIEKS and SCHAUB 22 Quintana-Domeque & Ródenas-Serrano,2017). 2 Second, our rich and fine-grained data makes it possible to apply a variety of econometric methods to more convincingly approximate causal estimates. Third, we explore potential direct and indirect pathways from terrorism to child mortality. Finally, we complement the empirical literature on the role of large-scale civil conflict (civil war) in child health. 3 We believe that our focus on terrorism is worthwhile because of the substantial differences between terrorism and other forms of larger-scale armed conflict. Compared to interstate and civil wars, the level of violence associated with terrorism tends to be very low (Gaibulloev & Sandler,2019). 4 Moreover, unlike rebel or regular armies, terrorist groups do not typically control territory, meaning that terrorism tends to create more diffused and punctuated threats rather than the objectively high threat levels affecting large areas associated with civil and interstate wars (Sambanis,2008). These differences imply that the direct effects of violence on child mortality—due to direct targeting or the destruction of health infrastructure, for example,—ought to be much more limited. Instead, whatever effect we detect is likely driven by indirect effects stemming from behavioral responses to terrorist violence. Employing a two-way fixed-effects approach, we show that higher levels of terrorist activity—operationalized as a terrorism index that accounts for both the frequency and intensity of terrorism—is associated with higher levels of child mortality. Using a panel event-study approach and a complimentary instrumental-variable (IV) approach that help to alleviate concerns about causal identification, we come to the same conclusion. The estimated effects are economically substantive, suggesting that plausible increases in terrorist activity are linked to several thousand additional deaths of children under the age of five per year. Interrogating our data, we show that the direct impact of terrorism (e.g., in terms of its lethality and destruction of public health infrastructure) tends to be small. This, in turn, suggests that increases in child mortality primarily emerge through the behavioral response of economic agents (e.g., parents, doctors, medical staff, aid workers and policymakers) to terrorism. Indeed, we provide tentative evidence that higher levels of terrorist activity unfavorably correlate with several proximate causes of child mortality such as the incidence of malaria and diarrhea, vaccination rates and malnourishment. The rest of this paper is organized as follows. Section2 discusses potential links between terrorism and child mortality and develops a testable hypothesis. We introduce the data on child mortality and terrorism in Section3. In Sections4 and 5, we empirically examine the terrorism-child mortality nexus. In Section6, we explore potential transmission channels from terrorism to child health. Section7 concludes. 2 | DIRECT AND INDIRECT EFFECTS OF TERRORISM ON CHILD MORTALITY 2.1 | Direct effects Most obviously, terrorism can adversely affect child mortality when children are killed in a terrorist attack. What is more, children may be wounded in an attack in ways that are eventually lethal. Similarly, terrorism may kill or incapacitate the children's parents, doctors and other medical personnel or foreign aid workers. This, in turn, may also contribute to child mortality by denying children parental or medical care. Finally, terrorism may destroy public health infrastructure (e.g., hospitals), which would likewise have direct adverse consequences for children's health. Still, while the direct effects of terrorism through the destruction of human life and the health infrastructure are eminently plausible, we do not expect them to affect child mortality in noticeable ways. This is because terrorism does not produce many victims, especially in comparison to many other sources of death. For instance, Arce(2019) estimates terrorism to lie in the bottom nine percent of the global burden of disease. That is, its burden is similar to that of Dengue fever and Vitamin A deficiency (Arce,2019,p.390). The direct impact of terrorism is also minor compared to other sources of violent death. For instance, for the early 2010s, Kamprad and Liem(2021) report that globally there were roughly half a million deaths per year from homicide (implying a homicide rate of 6.2 per 100,000 individuals), while terrorism accounted for approximately 38,000 deaths per year (implying a terrorism casualty rate of approximately 0.5 per 100,000 individuals). 2.2 | Indirect effects It is more probable that the adverse consequences of terrorism for child mortality are due to its indirect effects. These indirect effects emerge from the behavioral response of a variety of economic agents (e.g., parents, especially mothers; doctors and other healthcare workers; the government) to terrorism. This response, in turn, affects both the demand for and supply of children's healthcare in ways that increase the risk of child mortality. The parental perspective concerns the demand for children's healthcare. Here, we expect parents to be intimidated by terrorism. Indeed, the production of fear and intimidation for political leverage is a major goal of terrorist organizations (Gaibulloev MEIERRIEKS and SCHAUB 23 & Sandler,2019). What is more, Sunstein(2003) stresses the role of probability neglect, where individuals focus on a bad outcome (in our case, being harmed by a terrorist attack) but do not consider that this outcome is very unlikely to occur (as terrorism is very rare). Sunstein(2003) argues that the probability of harm is especially likely to be neglected when people's emotions are activated such as when their children's lives are threatened. The interplay between fear and probability neglect is consequently expected to lead to a behavioral response to terrorism on the part of affected parents that is potentially excessive. For instance, parents may forego preventive care (e.g., vaccinations or regular check-ups) out of fear that their children will be victimized by terrorism—a behavior that is expected to eventually contribute to higher levels of child mortality (e.g., Druetz etal.,2020; Rodríguez,2022). That is, the interaction between fear of terrorism and probability neglect may lead parents to weigh the risk of their child being harmed by terrorism more strongly than the child's risk of being affected by infectious diseases or other preventable causes of harm, even though this is not warranted given the actual probabilities of suffering harm. An stronger-than-warranted behavioral response due to fear of terrorism may also affect the supply of children's healthcare. Doctors, medical staff and international aid workers may stay at home rather than go to work, which could contribute to adverse consequences for children's health. At the same time, however, one could argue that the role of fear as a driving force of behavioral change in response to terrorism is less relevant for medical and aid workers as their professional experience may make them less vulnerable to emotional shocks. However, we can still hypothesize about behavioral responses to terrorism by medical and aid workers by considering a rational-choice perspective. Indeed, a number of theoretical contributions apply this perspective to provide an economic analysis of terrorism (e.g., Becker & Rubinstein, 2011; Eckstein & Tsiddon,2004; Naor, 2006; Sandler & Enders,2004; Schneider etal.,2015). The rational-choice approach posits that individuals are utility-maximizers, and that terrorism affects the individual utility-maximization process by influencing the costs and benefits associated with certain activities, potentially effecting behavioral changes when new utility-maximizing choices emerge. For one, terrorism is expected to reduce the benefits associated with certain choices of action of doctors, medical staff and international aid workers. For instance, it may reduce the benefits of work when parents do not come to see doctor and thus do not pay for the doctor's services. For another, terrorism may impose additional costs on medical and aid workers, for example, by necessitating additional investment into personal security. Ceteris paribus, the advent of terrorism is thus expected to make activities that may involve encountering terrorism (e.g., providing aid to or working and making patient visits in terror-ridden areas) less attractive. Instead, doctors, medical staff and international aid workers are expected to opt for those activities that avoid terrorism as those activities are now more likely to maximize utility. For instance, high-skilled doctors and medical workers may migrate away from terror-affected areas (e.g., Dreher etal.,2011). Such behavioral adjustments that follow from utility-maximization considerations could eventually result in poorer supply of child healthcare and higher levels of child mortality. Finally, terrorism may adversely affect the supply of healthcare by influencing public spending decisions. For one, the threat of terrorism may lead to increased public spending on security (e.g., Cevik & Ricco,2020; Gupta etal.,2004). We can explain this shift in spending by political considerations, where policymakers offer voters (who are intimidated by terrorism) security spending as a solution to the terrorist threat; by satisfying public demand for security in this manner, policymakers hope to maximize voter support. Alternatively, one may argue that the government consciously withdraws public spending from terror-ridden areas as a form of punishment when it suspects the local population to support terrorist activity. In any event, a shift or withdrawal of spending may come at the expense of public health expenditure, especially when resources are scarce in the first place. Because of terrorism-induced cuts in spendingthere may be fewer resources available for clinics, doctors, prevention and vaccination programs or health education. This lack of funding, in turn, is expected to adversely affect child health. 2.3 | Main hypothesis In line with our discussion, our main hypothesis is as follows: Higher levels of terrorist activity result in a higher risk of child mortality. In detail, terrorism may adversely affect children's health, first,through the destruction of human life and the health infrastructure (direct effects) and, second, due to the response of economic agents that undermines the adequate demand for and supply of children's healthcare (indirect effects). The behavioral response to terrorism may be related to psychological (fear and probability neglect), economic-rational (utility maximization) and political (public support and vote maximization) mechanisms. Given that terrorism's destructiveness tends to be comparatively low, we expect the various indirect consequences of terrorism to be the main reason for the hypothesized detrimental effect of terrorism on children's health. MEIERRIEKS and SCHAUB 24 3 | DATA To test our main hypothesis, we use sub-national data aggregated at the 0.5×0.5° (∼55×55km at the equator) grid-year level using the PRIO-GRID (Tollefsen etal.,2012) for a maximum of 52 African countries and territories for the 2000–2017 period. 5 The summary statistics for all variables employed in our analysis are reported in Supplementary Table1. 3.1 | Measuring child health outcomes Our main outcome of interest is child mortality, measured as the probability for a given child to die before reaching the age of five. The data comes from Burstein etal.(2019) who provide high-resolution (5km 2) estimates for lowand middle-income countries covering the whole of mainland Africa and Madagascar for the 2000–2017 period. Their geospatial estimates are derived from the collection of available Demographic and Health Surveys, UNICEF Multiple Indicator Cluster Surveys and other country-specific surveys. 6 We aggregate their data to the PRIO-GRID level. As part of our robustness checks, we also use two alternative child mortality measures from Burstein etal.(2019), neo-natal mortality (i.e., the risk of death for a new-born in the first 28days after birth) and infant mortality (i.e., the mortality risk under the age of one). As shown in Figure1, regardless of which indicator we choose, mortality rates generally saw a noticeable decline over our period of observation. For instance, the average risk of death for children under the age of five was 12.8% in the year 2000 but fell to approximately seven percent in the year 2017. For our subsequent analysis, this implies that estimating the effect of terrorism on mortality outcomes primarily means assessing whether terrorism produced conspicuous setbacks from the general downward trend in mortality. 3.2 | Measuring terrorist activity Our main independent variable is an index of terrorist activity. Similar to the index of Eckstein and Tsiddon(2004), it is defined as the sum of the per capita number of terrorist attacks and per capita number of terrorism casualties per grid-year observation. The term “terrorism casualties” refers to the number of individuals that are killed in a terrorist attack. The index reflects both the frequency (number of terrorist attacks) and ferocity (number of casualties) of terrorism. To reduce the influence of outliers, we apply the inverse hyperbolic sine transformation to our terrorism index. 7 In weighing our terrorism variable by population size, we follow Jetter and Stadelmann(2019). They suggest that per capita measures of terrorism are more reflective of the (individual) risk associated with terrorism. As stressed above, we argue that it is this very risk that explains how terrorism may (indirectly) affect child mortality. Similar population-adjusted terrorism indicators are used in Tavares(2004), Gaibulloev and Sandler(2011) and Meierrieks and Gries(2013). FIGURE 1 Mortality rates, 2000–2017. MEIERRIEKS and SCHAUB 25 The data on grid-level population size (used to calculate the per capita rates) come from the LandScan high-resolution global population data set (Bright etal.,2018). The terrorism data are from the Global Terrorism Database (GTD). The GTD was first described in LaFree and Dugan(2007). It collects information on terrorist activity from reputable media outlets. 8 For a terrorist event to be recorded in the GTD, it must be documented by at least one high-quality media source (e.g., a renowned international newspaper such as the New York Times). To be considered a terrorist event, it must also (1) be intentional, (2) entail some level of violence or threat of violence and (3) be committed by non-state actors, meaning that violence by state actors is excluded (LaFree & Dugan,2007). Furthermore, it must meet at least two of the following three criteria: (1) the incident must be carried out to achieve a political, economic, religious or social goal, (2) there must be evidence of an intention to coerce, intimidate or convey some other message to a larger audience than the immediate victims and/or (3) the incident must be outside the context of conventional warfare (LaFree & Dugan,2007). The GTD provides geolocational information (latitude and longitude) on 99% of all attacks reported. This allows us to combine the terrorism index with our health data. To add to the robustness of our findings, below we also use the constitutive partsof our terrorism index as explanatory variables, that is, the (inverse hyperbolic sine transformed) per capita number of terrorist incidents and the (inverse hyperbolic sine transformed) per capita number of terrorism casualties per grid-year observation. This is to assess whether any effect of terrorism on child mortality is due to the frequency or the ferocity of terrorism. Note, however, that the casualty variable is likely subject to under-counting as terrorism victim figures are unknown for many observations in the GTD. Figure2 visualizes the temporal trends in terrorism (indicated by the annual mean of the terrorism index) in Africa over our period of observation. There is a clear uptick in terrorist activity after 2011. While there were, on average, approximately 310 terrorist attacks per year between 2000 and 2011, the annual average was almost 2350 attacks from 2012 onwards. This increase can partly be attributed to stronger Islamist terrorist activity in the 2010s, for example, to attacks by Al-Qaida in the Islamic Maghreb in Algeria and Mali as well as Boko Haram in Nigeria, Niger and Chad. Furthermore, terrorism in Africa is linked to violent separatism such as in Ethiopia (e.g., by the Oromo Liberation Front) and Angola (e.g., by the Front for the Liberation of the Enclave of Cabinda). 3.3 | Geography of child mortality and terrorism We illustrate the geographical distribution of child mortality and terrorism in Figure3. The strength of the shading indicates the severity of child mortality, measured as the probability of a child to die before reaching the age of five, averaged over the 2000–2017 period and aggregated at the grid level. Figure3 shows that child mortality rates are much higher in the Sahel and Central Africa compared to Southern and Northern Africa. Black dots indicate the location of individual terrorist attacks between 2000 and 2017. Some countries (e.g., the Comoros, Lesotho and Morocco) were almost completely unaffected by terrorism, while other countries (e.g., Algeria, Egypt and Nigeria) saw substantial terrorist activity. Finally, Figure3 also points to a strong intra-country heterogeneity both with respect to child mortality and terrorism. For instance, the northern and south-eastern parts of Nigeria saw both markedly higher child mortality and more terrorist attacks than the rest of the country. FIGURE 2 Terrorism over time. MEIERRIEKS and SCHAUB 26 4 | TWO-WAY FIXED-EFFECTS APPROACH 4.1 | Empirical model To study the relationship between children's health and terrorist activity, we consider the following two-way fixed-effects model, which we estimate using the OLS-estimator: ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒ℎ𝑘𝑘𝑘𝑘𝑘𝑒𝑒 =𝛽𝛽1∗𝑒𝑒𝑒𝑒𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑗𝑗𝑘𝑘𝑘𝑒𝑒 +𝛽𝛽2∗𝑋𝑋𝑘𝑘𝑒𝑒 +𝛼𝛼𝑘𝑘+𝜆𝜆𝑒𝑒+𝑣𝑣𝑘𝑘𝑒𝑒 (1) Here, health refers to our kth measure of children's heath in grid i and yeart. Usually, this is the risk of child mortality for children under the age of five, but it may also indicate the neo-natal or infant mortality risk. The variable terror refers to our jth measure of terrorist activity. Commonly, we employ our terrorism index but as a robustness check we also use two alternative terrorism indicators. We include grid-specific fixed-effects (α) to control for the role of time-invariant factors that may confound the relationship between terrorism and children's health. For instance, local geographical conditions (e.g., proximity to mosquito habitats) may be conducive to the spread of malaria, which, in turn, adversely affects children's health. At the same time, certain geographical conditions (e.g., proximity to forested areas) may provide potential militants with safe havens and thus facilitate terrorist activity (Fearon & Laitin, 2003; Schaub & Auer, 2023). Similarly, year-fixed effects (λ) control for the influence of global trends and events that may have affected terrorism and public health. For instance, the introduction or diffusion of medical technology during our period of observation may have reduced child mortality. Also, global trends in terrorism (e.g., the rise of Al-Qaida and the Islamic State) ought to have influenced terrorism in Africa. Finally, we account for a set of time-variant confounders (X) that may affect both public health and terrorist activity and thus also obfuscate—due to omitted variable bias—the relationship between terrorism and child mortality. Because data on the controls is not available for all grids and years, our sample size is reduced when including them. When considering only the role FIGURE 3 Geography of terrorism and child mortality. [Colour figure can be viewed at wileyonlinelibrary.com] MEIERRIEKS and SCHAUB 27 of terrorism (plus the fixed effects) in child mortality in a parsimonious model, our sample covers 7954 grids for the 2000–2017 period. A model that also considers the role of the various confounders allows us to draw on 6751 grids for the 2000–2015 period. We control for (1) nightlights (i.e., light emissions during nighttime) as a measure of economic development, using data from Elvidge etal.(1997, 2021) and Ghosh etal.(2021); (2) travel time to the nearest large city as a measure of quality of local infrastructure, drawing data from Müller-Crepon(2021); (3) urban population, where the data are from an update of Meiyappan and Jain(2012); (4) female education, measured as the years of education of 20-24-year-old women, where the data are from Graetz etal.(2018); and (5) temperature (in °C) and precipitation (measured by the standardized precipitation evapotranspiration index), where the data are from Fan and van den Dool(2008) and Peng etal.(2019). We apply the inverse hyperbolic sine transformation to all controls except for the climate variables to correct for skewness. All controls are expected to affect both the prevalence of terrorism and child health outcomes (Brockhoff etal.,2015; Craig etal.,2021; Eckert & Kohler,2014; Freytag etal.,2011; Hahn etal.,2018; Kis-Katos etal.,2014; Meierrieks,2021; Sheffield & Landrigan,2011; Subbarao & Raney,1995). For instance, we expect richer grids to see lower levels of child mortality (e.g., by virtue of better access to medical technology); at the same time, economic development may also affect terrorism, for example, by influencing its opportunity costs (Freytag etal.,2011; Kis-Katos etal.,2014). 4.2 | Empirical results We report our findings in Table1. We find that more terrorist activity is associated with higher levels of child mortality. This main finding is robust to the inclusion of the baseline controls. Considering economic substantiveness, our baseline estimates (Model (3), Table1) suggest that an increase in terrorism by 10 percent is associated with an increase in child mortality of approximately 0.01% points. To give a comparison, an increase in economic development (nightlights) by 10 percent will yield a decrease in child mortality of approximately 0.02% points. Our main finding is also robust to different measurements of mortality and terrorism. Concerning the latter, the association between the frequency of terrorism and child mortality appears to be somewhat stronger than the association between the lethality of terrorism and child mortality. Potentially, the ferocity of terrorism is more strongly clouded by uncertainty and underreporting, thus making behavioral responses to it by affected economic agents less straightforward. Finally, allowing for a more complex lag structure with respect to the correlation between terrorism and the risk of child mortality suggests that this association becomes somewhat stronger over time. 9 This may correspond to behavioral changes in response to terrorism that take some time to materialize. For example, this may pertain to a migratory response to terrorism on the part of medical workers, which, in turn, will only impact child mortality in subsequent years. 4.3 | Role of civil conflict and war Above, we noted that terrorism is distinct from large-scale violence, for example, with respect to the number of casualties, the targeting of civilians and the control of territory by non-state actors (e.g., Sambanis,2008). These differences motivated our empirical approach to study the terrorism-child mortality nexus. At the same time, however, one may point to potential overlaps between terrorist activity and larger-scale civil conflict (e.g., Findley & Young,2012). 10 If there is a systematic association between terrorism and larger-scale civil conflict, then an empirical investigation of the effect of terrorism on child health may also pick up the clearly established impact of civil conflict (e.g., Bendavid etal.,2021), thus over-estimating the impact of terrorism on child mortality. To study this latter proposition, from the UCDP Georeferenced Event Dataset (Sundberg & Melander,2013) we extract the number of deaths due to the use of armed force by an organized actor against another organized actors per grid-year observation. Analogous to our terrorism index, this death count is weighted by local population size and the inverse hyperbolic sine transformation is applied to it. These deaths will not be considered by the GTD as the GTD does not include activity by combatants (i.e., violence between organized actors such as the government against rebel groups). At the same time, our UCDP-based measure explicitly excludes violence against civilians because this might be included as terrorism by the GTD. Consequently, we are confident that using the UCDP and GTD data allows us to differentiate between small-scale terrorism (GTD) and large-scale civil conflict and war (UCDP). Indeed, the average number of victims per grid-year observation due to terrorism is 0.3, while it is 1.7 for civil conflict. The correlation between terrorism and civil conflict victims is positive but small (r=0.04, p<0.01). MEIERRIEKS and SCHAUB 28 For our empirical analysis of the role of civil conflict, in addition to the (inverse hyperbolic sine transformed) per capita number of battle deaths, we also consider a binary measure that is equal to unity when there is at least one battle death and a dichotomous measure of civil war that is equal to unity when a grid observation sees at least 25 battle deaths per year (which is a common civil war death threshold; see Blattman & Miguel,2010). In addition, we run two models where we drop all grid-year observations when they see at least one battle death or all grids with any battle death during our period of observation (even when they experience years without battle deaths), respectively. As shown in Supplementary Table2, we find that the positive association between terrorism and child mortality holds after controlling for various measures of civil conflict/war and when excluding civil conflict grids and episodes. Here, the point estimates concerning the association between terrorism and mortality are—as expected—somewhat smaller but still reasonably close to our baseline results reported in Table1. Our results suggest that increases in both the scope of terrorism and civil conflict have a comparable association with child mortality. For instance, a 10 percent increase in the terrorism index and a 10 percent increase in the number of battle deaths are both expected to increase the child mortality risk by approximately 0.01% points. The latter positive association between civil conflict and child mortality is consistent with earlier findings such as Bendavid etal.(2021). At the same time, the results reported in Supplementary Table2 indicate that terrorism shares an unfavorable relationship with child mortality that is independent of the impact of larger-scale conflicts. (1) (2) (3) (4) (5) (6) (7) Dependent variable → Mort. U5 Mort. U5 Mort. U5 Mort. Neo Mort. U1 Mort. U5 Mort. U5 Mort. U5 Terrorism index 0.048*** 0.088*** 0.087*** 0.014** 0.058*** 0.028* (0.016) (0.020) (0.019) (0.005) (0.011) (0.015) Terrorism index t-1 0.071*** (0.015) Terrorism index t-2 0.125*** (0.015) [Sum of coefficients] [0.225]*** [Standard error] [0.037] Terrorist incidents p.c. 0.167*** (0.030) Terrorism casualties p.c. 0.071*** (0.021) Nightlights −0.225** −0.163*** −0.219*** −0.225** −0.226** −0.133 (0.106) (0.024) (0.053) (0.107) (0.107) (0.089) Distance to city 0.411*** 0.170*** 0.120** 0.413*** 0.411*** 0.266*** (0.099) (0.027) (0.055) (0.099) (0.099) (0.094) Urban −1.778*** −0.326** −0.560 −1.767*** −1.775*** −1.918*** (0.649) (0.149) (0.342) (0.649) (0.649) (0.644) Female education −0.443*** −0.145*** −0.190*** −0.444*** −0.442*** −0.275*** (0.038) (0.011) (0.022) (0.038) (0.038) (0.042) Temperature 0.023** −0.001 0.007 0.023** 0.023*** 0.021*** (0.009) (0.002) (0.004) (0.009) (0.009) (0.008) Precipitation (SPEI) 0.003 0.004*** 0.001 0.003 0.003 0.019*** (0.003) (0.001) (0.002) (0.003) (0.003) (0.003) Observations 142,473 101,089 101,089 101,089 101,089 101,089 101,089 87,528 Number of grids/clusters 7954 6751 6751 6751 6751 6751 6751 6746 Adjusted R 20.95 0.96 0.96 0.95 0.95 0.96 0.96 0.96 Note: Mort. U5=Mortality risk of children under the age of five. Mort. Neo=Mortality risk of children in first 28days after birth. Mort. U1=Mortality risk of children under the age of one. Gridand year-fixed effects always included. Standard errors clustered at the grid-level in parentheses. *p<0.1, **p<0.05, ***p<0.01. TABLE 1 Two-way fixed-effects estimates. MEIERRIEKS and SCHAUB 35 Panel B: Correlation between potential mediators and child mortality (1b) (2b) (3b) (4b) Mediator variable → Malaria Diarrhea Vaccinations Malnutrition Mediator 1.173*** 1.659*** −1.481*** 6.950*** (0.071) (0.100) (0.129) (0.446) Nightlights −0.360*** −0.233** −0.258** −0.312*** (0.106) (0.104) (0.107) (0.101) Distance to city 0.535*** 0.308*** 0.516*** 0.417*** (0.099) (0.099) (0.099) (0.095) Urban −1.996*** −1.490** −1.419** −2.006*** (0.660) (0.634) (0.631) (0.638) Female education −0.443*** −0.369*** −0.409*** −0.210*** (0.038) (0.039) (0.038) (0.037) Temperature 0.011 0.023*** 0.032*** 0.015* (0.009) (0.009) (0.009) (0.009) Precipitation (SPEI) 0.002 0.003 −0.000 0.002 (0.004) (0.003) (0.003) (0.003) Adjusted R 20.96 0.95 0.96 0.96 Number of grids/clusters 6677 6751 6751 6750 Observations (both panels) 100,006 101,089 101,074 101,074 Note : OLS-estimates reported. Dependent variable in Panel B=Mortality risk of children under the age of five. Gridand year-fixed effects always included. Standard errors clustered at the grid-level in parentheses. *p<0.1, **p<0.05, ***p<0.01. TABLE 2 (Continued) MEIERRIEKS and SCHAUB 36 According to these back-of-the-envelope calculations, our estimates thus point to substantial annual increases in child mortality due to terrorist activity especially when the scale of escalation is substantial and long-lasting. To this effect, our findings also speak to the existing literature on the adverse consequences of large-scale conflict on child health. Given that the direct impact of terrorism (in terms of its lethality and destruction of public health infrastructure) tends to be very small, we are reasonably confident that increases in child mortality primarily emerge through the behavioral response of economic agents (e.g., parents, doctors, medical staff, aid workers and policymakers) to terrorism. Exploring the role of several potential mediators in the terrorism-child mortality nexus, we show that higher levels of terrorist activity unfavorably correlate with several proximate causes of child mortality: the incidence of malaria and diarrhea, vaccination rates and malnourishment. Future work could examine these behavioral responses and mediators in more detail. For instance, it could be interesting to examine—using appropriate micro-level or survey data—whether it is parents, doctors, politicians or other economic agents that respond especially unfavorably (in terms of the consequences for child health) to terrorism. Recent years saw encouraging advances in reducing child mortality in Africa. Our empirical analysis, however, suggests that terrorism produced some conspicuous setbacks with respect to this trend. Domestic and international policymakers are thus called upon to counter terrorism but should also consider mitigating behavioral responses to terrorism that are to the detriment of children's health, for example, through information and education campaigns that adjust perceptions about terrorism and the risk it entails. ACKNOWLEDGMENTS We would like to thank Daniel Arce, Janina Beiser-McGrath, Anke Hoeffler, Oguzhan Turkoglu as well as the participants of EPSA 2022, the Jan Tinbergen European Peace Science Conference 2022, the 2023 Annual Meeting of the European Public Choice Society and the MACIE seminar at the University of Marburg for their valuable feedback. The data used in this study are publicly available and no specific funding was acquired. The instructions to access the data and the code used in the analyses will be made available on Harvard's Dataverse. Open Access funding enabled and organized by Projekt DEAL. CONFLICT OF INTEREST STATEMENT Both authors contributed equally to this work and declare no conflict of interests. DATA AVAILABILITY STATEMENT The data and the code used in the analyses will be made available on Harvard Dataverse. ORCID Daniel Meierrieks https://orcid.org/0000-0003-2058-8385 Max Schaub https://orcid.org/0000-0003-2057-7002 ENDNOTES 1 Terrorism is “the premeditated use or threat to use violence against noncombatants by individuals or subnational groups to obtain a political objective through the intimidation of a large audience beyond that of the immediate victims” (Gaibulloev & Sandler,2019,p.278). 2 Furthermore, by highlighting the role of terrorism in child health and mortality, we also add to the larger empirical literature on the socio-economic consequences of terrorism (e.g., Tavares,2004; Frey etal.,2007; Gaibulloev & Sandler,2011; Meierrieks & Gries,2013; Kim & Albert Kim,2018; Blasco etal.,2022; for an overview see Gaibulloev & Sandler,2019). 3 Beside its unfavorable direct effects, child mortality may increase as a consequence of large-scale conflict because of the spread of infectious diseases (e.g., Charchuk etal.,2016; Iqbal & Zorn,2010), in utero exposure to conflict (Akbulut-Yuksel,2017; Aparicio Fenoll & González,2021; Dagnelie etal.,2018), the destruction of health infrastructure and the flight of health workers from conflict-ridden areas (e.g., Chi etal.,2015; Chukwuma & Ekhator-Mobayode,2019; McKay,1998; Price & Bohara,2013; Sharara & Kanj,2014), the destruction of sanitation, waste and water treatment (Kirschner & Finaret,2021), the underfunding of healthcare institutions by the government in times of conflict (Gates etal.,2012; Iqbal,2006) and reduced food supply (e.g., Lin,2022), which, in turn, is expected to correlate with weight loss and stunting (Bendavid etal.,2021; Bundervoet etal.,2009; Dunn,2018; Kirschner & Finaret,2021; Wagner etal.,2018, 2019). See also Kadir etal.(2019) for a literature review on the consequences of armed conflict for child health and development. 4 A common definition of armed civil conflict involves at least 25 (or, more conservatively, at least 1000) battle-related deaths per year in a specific country (Blattman & Miguel,2010). The magnitude of violence associated with terrorism is commonly much smaller (Gaibulloev & Sandler,2019). 5 These countries and territories are Algeria, Angola, Benin, Botswana, Burkina Faso, Burundi, Cameroon, Central African Republic, Chad, Comoros, Congo, Côte d'Ivoire, Democratic Republic of the Congo, Djibouti, Egypt, Equatorial Guinea, Eritrea, eSwatini, Ethiopia, Gabon, Gambia, MEIERRIEKS and SCHAUB 37 Ghana, Guinea, Guinea-Bissau, Kenya, Lesotho, Liberia, Libya, Madagascar, Malawi, Mali, Mauritania, Morocco, Mozambique, Namibia, Niger, Nigeria, Rwanda, South Sudan, Senegal, Sierra Leone, Somalia, Somaliland, South Africa, Sudan, Tanzania, Togo, Tunisia, Uganda, Western Sahara, Zambia and Zimbabwe. 6 The use of (self-reported) survey data to create the child mortality variable may lead to measurement error in this variable. As we use the child mortality variable as the dependent variable in our empirical analyses, we expect this potential measurement error to mainly affect the precision of our estimates but not to cause bias when estimating the effect of terrorism on this variable. Fortunately, since the number of observations available for our analyses is very large, this is anticipated to counteract losses of estimation precision due to measurement error in the dependent variable. 7 In contrast to the log trans-formation, the inverse hyperbolic sine transformation is also defined for grid-year observations with no terrorist activity (e.g., Burbidge etal.,1988; Bellemare & Wichman,2020). Note that as part of our robustness checks, we also use alternative transformations and operationalizations of our terrorism index. 8 The GTD can be accessed at https://www.start.umd.edu/gtd/. 9 Allowing for further lags yields similar findings. Controlling for a maximum of 10 lags, the terrorism index will now longer share a statistically significant association with child mortality after the sixth lag. Note, however, that because a more complex lag structure reduces our sample size, these results cannot be directly compared to our other estimates reported in Table1. 10 This overlap implies that while “terrorism may be used by rebel forces, terrorism need not be associated with civil war” (Gaibulloev & Sandler,2019,p.291). 11 We only examine the effect of terrorism on child mortality within the same grid cell. We consider (via the named standard errors) but do not explicitly operationalize spillover effects of terrorism in neighboring cells. Assessing the role of spatially indirect effects of terrorism on child mortality may be a fruitful avenue for future research. 12 As a robustness check, we also construct the sup-t confidence bands of Montiel Olea and Plagborg-Møller(2019) which are especially appropriate when we are interested in the entire event-time path, that is, in implicitly testing multiple hypotheses at once (Freyaldenhoven etal.,2021). Reassuringly, these confidence bands are like those in Figure4 to6. 13 The data can be found at https://population.un.org/wpp/. 14 Both figures are averages over our observation period. Due to population growth, total births are substantially higher in 2017 compared to 2000, while the child mortality risk (e.g., due to medical advances) is considerably lower in 2017 compared to 2000. 15 All outcome measures are made available by the Institute for Health Metrics and Evaluation on their website https://ghdx.healthdata.org/ local-and-small-area-estimation. 16 For instance, holding constant the number of terrorism casualties, such an increase would imply an increase of the Africa-wide absolute number of terrorist incidents by approximately 625. While such an escalation in violence is substantial, it was indeed observed for our sample between both 2011 and 2012 as well as between 2013 and 2014. 17 Yet, other causes of child mortality in this year were much more important. For instance, ca. 530,000 children died from malaria, ca. 425,000 from diarrhea, and ca. 600,000 from pneumonia in 2010 (Liu etal.,2012). REFERENCES Akbulut-Yuksel, M. (2017). War during childhood: The long run effects of warfare on health. Journal of Health Economics, 53(May), 117–130. https://doi.org/10.1016/j.jhealeco.2017.02.005 Aparicio Fenoll, A., & González, L. (2021). Political instability and birth outcomes: Evidence from the 1981 military coup in Spain. Health Economics, 30(2), 328–341. https://doi.org/10.1002/hec.4201 Arce, D. G. (2019). On the human consequences of terrorism. Public Choice, 178(3), 371–396. https://doi.org/10.1007/s11127-018-0590-9 Bakker, R., Hill, D. W., & Moore, W. H. (2016). How much terror? Dissidents, governments, institutions, and the cross-national study of terror attacks. Journal of Peace Research, 53(5), 711–726. https://doi.org/10.1177/0022343316655020 Becker, G., & Rubinstein, Y. (2011). Fear and the response to terrorism: An economic analysis. CEP Discussion Paper. Centre for Economic Performance, LSE. Retrieved from https://econpapers.repec.org/paper/cepcepdps/dp1079.htm Bellemare, M. F., & Wichman, C. J. (2020). Elasticities and the inverse hyperbolic sine transformation. Oxford Bulletin of Economics and Statistics, 82(1), 50–61. https://doi.org/10.1111/obes.12325 Bendavid, E., Boerma, T., Akseer, N., Langer, A., Malembaka, E. B., Okiro, E. A., Wise, P.H., Heft-Neal, S., Black, R. E., Bhutta, Z. A., Bhutta, Z., Black, R., Blanchet, K., Boerma, T., Gaffey, M., Spiegel, P., Waldman, R., & Wise, P. (2021). The effects of armed conflict on the health of women and children. The Lancet, 397(10273), 522–532. https://doi.org/10.1016/S0140-6736(21)00131-8 Blasco, S., Moreno-Galbis, E., & Tanguy, J. (2022). Getting used to terrorist threats? Evidence from French terrorist attacks between 2015 and 2016. Health Economics, 31(3), 508–540. https://doi.org/10.1002/hec.4465 Blattman, C., & Miguel, E. (2010). Civil war. Journal of Economic Literature, 48(1), 3–57. https://doi.org/10.1257/jel.48.1.3 Bright, E. A., Rose, A. N., Urban, M. L., & Jacob, McK. (2018). LandScan 2017 high-resolution global population data set. Oak Ridge National Lab. (ORNL). Brockhoff, S., Krieger, T., & Meierrieks, D. (2015). Great expectations and hard times: The (nontrivial) impact of education on domestic terrorism. Journal of Conflict Resolution, 59(7), 1186–1215. https://doi.org/10.1177/0022002713520589 MEIERRIEKS and SCHAUB 38 Bundervoet, T., Verwimp, P., & Akresh, R. (2009). Health and civil war in rural Burundi. Journal of Human Resources, 44(2), 536–563. https://doi. org/10.3368/jhr.44.2.536 Burbidge, J. B., Magee, L., & eRobb, A. L. (1988). Alternative transformations to handle extreme values of the dependent variable. Journal of the American Statistical Association, 83(401), 123–127. https://doi.org/10.1080/01621459.1988.10478575 Burstein, R., Henry, N. J., Collison, M. L., Marczak, L. B., Sligar, A., Watson, S., Neal, M., Abbasalizad-Farhangi, M., Abbasi, M., Abd-Allah, F., Abdoli, A., Abdollahi, M., Abdollahpour, I., Abdulkader, R. S., Abrigo, M. R. M., Acharya, D., Adebayo, O. M., Adekanmbi, V., Adham, D. & Hay, S. I. (2019). Mapping 123 million neonatal, infant and child deaths between 2000 and 2017. Nature, 574(7778), 353–358. https://doi. org/10.1038/s41586-019-1545-0 Camacho, A. (2008). Stress and birth weight: Evidence from terrorist attacks. The American Economic Review, 98(2), 511–515. https://doi. org/10.1257/aer.98.2.511 Cevik, S., & Ricco, J. (2020). Shock and awe? Fiscal consequences of terrorism. Empirical Economics, 58(2), 723–748. https://doi.org/10.1007/ s00181-018-1543-3 Charchuk, R., Paul, M. K. J., Masumbuko Claude, K., Houston, S., & Hawkes, M. T. (2016). Burden of malaria is higher among children in an internal displacement camp compared to a neighbouring village in the democratic republic of the Congo. Malaria Journal, 15(1), 431. https://doi. org/10.1186/s12936-016-1479-z Chi, P.C., Bulage, P., Urdal, H., & Sundby, J. (2015). Perceptions of the effects of armed conflict on maternal and reproductive health services and outcomes in Burundi and northern Uganda: A qualitative study. BMC International Health and Human Rights, 15(1), 7. https://doi.org/10.1186/ s12914-015-0045-z Chukwuma, A., & Ekhator-Mobayode, U. E. (2019). Armed conflict and maternal health care utilization: Evidence from the Boko Haram insurgency in Nigeria. Social Science & Medicine, 226(April), 104–112. https://doi.org/10.1016/j.socscimed.2019.02.055 Clarke, D., & Tapia-Schythe, K. (2021). Implementing the panel event study. STATA Journal, 21(4), 853–884. https://doi.org/10.1177/1536867X211063144 Conley, T. G., Hansen, C. B., & Rossi, P.E. (2012). Plausibly exogenous. The Review of Economics and Statistics, 94(1), 260–272. https://doi. org/10.1162/REST_a_00139 Craig, C. M., Overbeek, R. W., & Niedbala, E. M. (2021). A global analysis of temperature, terrorist attacks, and fatalities. Studies in Conflict & Terrorism, 44(11), 958–970. https://doi.org/10.1080/1057610X.2019.1606992 Dagnelie, O., De Luca, G. D., & Maystadt, J.-F. (2018). Violence, selection and infant mortality in Congo. Journal of Health Economics, 59(May), 153–177. https://doi.org/10.1016/j.jhealeco.2018.02.004 Dreher, A., Krieger, T., & Meierrieks, D. (2011). Hit and (they will) run: The impact of terrorism on migration. Economics Letters, 113(1), 42–46. https://doi.org/10.1016/j.econlet.2011.05.044 Druetz, T., Browne, L., Frank, B., Mitchell, M. I., & Abel, B. (2020). Effects of terrorist attacks on access to maternal healthcare services: A national longitudinal study in Burkina Faso. BMJ Global Health, 5(9), e002879. https://doi.org/10.1136/bmjgh-2020-002879 Dunn, G. (2018). The impact of the Boko Haram insurgency in northeast Nigeria on childhood wasting: A double-difference study. Conflict and Health, 12(1), 6. https://doi.org/10.1186/s13031-018-0136-2 Eckert, S., & Kohler, S. (2014). Urbanization and health in developing countries: A systematic review. World Health & Population, 15(1), 7–20. https://doi.org/10.12927/whp.2014.23722 Eckstein, Z., & Tsiddon, D. (2004). Macroeconomic consequences of terror: Theory and the case of Israel. Journal of Monetary Economics, 51(5), 971–1002. https://doi.org/10.1016/j.jmoneco.2004.05.001 Elvidge, C. D., Baugh, K. E., Kihn, E. A., Kroehl, H. W., & Davis, E. R. (1997). Mapping city lights with nighttime data from the DMSP operational linescan system. Photogrammetric Engineering & Remote Sensing, 63(6), 727–734. Elvidge, C. D., Zhizhin, M., Ghosh, T., Hsu, F.-C., & Taneja, J. (2021). Annual time series of global VIIRS nighttime lights derived from monthly averages: 2012 to 2019. Remote Sensing, 13(5), 922. https://doi.org/10.3390/rs13050922 Fan, Y., & van den Dool, H. (2008). A global monthly land surface air temperature analysis for 1948–present. Journal of Geophysical Research: Atmospheres, 113(D1), D01103. https://doi.org/10.1029/2007JD008470 Fearon, J. D., & Laitin, D. D. (2003). Ethnicity, insurgency, and civil war. American Political Science Review, 97(1), 75–90. Findley, M. G., & Young, J. K. (2012). Terrorism and civil war: A spatial and temporal approach to a conceptual problem. Perspectives on Politics, 10(2), 285–305. https://doi.org/10.1017/S1537592712000679 Frey, B. S., Luechinger, S., & Stutzer, A. (2007). Calculating tragedy: Assessing the costs of terrorism. Journal of Economic Surveys, 21(1), 1–24. https://doi.org/10.1111/j.1467-6419.2007.00505.x Freyaldenhoven, S., Hansen, C., Pérez, J. P., & Shapiro, J. M. (2021). Visualization, identification, and estimation in the linear panel event-study design. National Bureau of Economic Research. Working Paper 29170. Working Paper Series. https://doi.org/10.3386/w29170 Freytag, A., Krüger, J. J., Meierrieks, D., & Schneider, F. (2011). The origins of terrorism: Cross-country estimates of socio-economic determinants of terrorism. European Journal of Political Economy, 27(December), S5–S16. Special Issue:Terrorism. https://doi.org/10.1016/j. ejpoleco.2011.06.009 Gaibulloev, K., & Sandler, T. (2011). The adverse effect of transnational and domestic terrorism on growth in Africa. Journal of Peace Research, 48(3), 355–371. https://doi.org/10.1177/0022343310395798 Gaibulloev, K., & Sandler, T. (2019). What we have learned about terrorism since 9/11. Journal of Economic Literature, 57(2), 275–328. https://doi. org/10.1257/jel.20181444 Gates, S., Hegre, H., Nygård, H. M., & Strand, H. (2012). Development consequences of armed conflict. World Development, 40(9), 1713–1722. https://doi.org/10.1016/j.worlddev.2012.04.031 MEIERRIEKS and SCHAUB 39 Ghosh, T., Baugh, K. E., Elvidge, C. D., Zhizhin, M., Poyda, A., & Hsu, F.-C. (2021). Extending the DMSP nighttime lights time series beyond 2013. Remote Sensing, 13(24), 5004. https://doi.org/10.3390/rs13245004 Graetz, N., Friedman, J., Osgood-Zimmerman, A., Roy, B., Biehl, M. H., Shields, C., Mosser, J. F., Casey, D. C., Deshpande, A., Earl, L., Reiner, R. C., Ray, S. E., Fullman, N., Levine, A. J., Stubbs, R. W., Mayala, B. K., Longbottom, J., Browne, A. J., Bhatt, S., & Hay, S. I. (2018). Mapping local variation in educational attainment across Africa. Nature, 555(7694), 48–53. https://doi.org/10.1038/nature25761 Gupta, S., Clements, B., Bhattacharya, R., & Chakravarti, S. (2004). Fiscal consequences of armed conflict and terrorism in lowand middle-income countries. European Journal of Political Economy, 20(2), 403–421. The Economic Consequences of Terror. https://doi.org/10.1016/j. ejpoleco.2003.12.001 Hahn, Y., Nuzhat, K., & Yang, H.-S. (2018). The effect of female education on marital matches and child health in Bangladesh. Journal of Population Economics, 31(3), 915–936. https://doi.org/10.1007/s00148-017-0673-9 Iqbal, Z. (2006). Health and human security: The public health impact of violent conflict. International Studies Quarterly, 50(3), 631–649. https:// doi.org/10.1111/j.1468-2478.2006.00417.x Iqbal, Z., & Zorn, C. (2010). Violent conflict and the spread of HIV/AIDS in Africa. The Journal of Politics, 72(1), 149–162. https://doi.org/10.1017/ S0022381609990533 Jetter, M., & Stadelmann, D. (2019). Terror per capita. Southern Economic Journal, 86(1), 286–304. https://doi.org/10.1002/soej.12369 Kadir, A., Shenoda, S., & Goldhagen, J. (2019). Effects of armed conflict on child health and development: A systematic review. PLoS One, 14(1), e0210071. https://doi.org/10.1371/journal.pone.0210071 Kaila, H., Nawo, L., & Son, H. H. (2021). Unpacking the links between conflict and child welfare: Evidence from a foreign insurgency. Kamprad, A., & Liem, M. (2021). Terror and the legitimation of violence: A cross-national analysis on the relationship between terrorism and homicide rates. Terrorism and Political Violence, 33(1), 96–118. https://doi.org/10.1080/09546553.2018.1523150 Kim, D., & Albert Kim, Y.-I. 1 (2018). Mental health cost of terrorism: Study of the charlie hebdo attack in Paris. Health Economics, 27(1), e1–e14. https://doi.org/10.1002/hec.3520 Kirschner, S. A., & Finaret, A. B. (2021). Conflict and health: Building on the role of infrastructure. World Development, 146(October), 105570. https://doi.org/10.1016/j.worlddev.2021.105570 Kis-Katos, K., Liebert, H., & Schulze, G. G. (2014). On the heterogeneity of terror. European Economic Review, 68(C), 116–136. https://doi. org/10.1016/j.euroecorev.2014.02.009 LaFree, G., & Dugan, L. (2007). Introducing the global terrorism Database. Terrorism and Political Violence, 19(2), 181–204. https://doi. org/10.1080/09546550701246817 Lauderdale, D. S. (2006). Birth outcomes for Arabic-named women in California before and after September 11. Demography, 43(1), 185–201. https://doi.org/10.1353/dem.2006.0008 Lin, E. (2022). How war changes land: Soil fertility, unexploded bombs, and the underdevelopment of Cambodia. American Journal of Political Science, 66(1), 222–237. https://doi.org/10.1111/ajps.12577 Liu, Li, Johnson, H. L., Cousens, S., Perin, J., Scott, S., Lawn, J. E., Rudan, I., Campbell, H., Cibulskis, R., Li, M., Mathers, C., & Black, R. E. (2012). Global, regional, and national causes of child mortality: An updated systematic analysis for 2010 with time trends since 2000. Lancet (London, England), 379(9832), 2151–2161. https://doi.org/10.1016/S0140-6736(12)60560-1 Mansour, H., & Rees, D. I. (2012). Armed conflict and birth weight: Evidence from the al-aqsa intifada. Journal of Development Economics, 99(1), 190–199. https://doi.org/10.1016/j.jdeveco.2011.12.005 McKay, S. (1998). The effects of armed conflict on girls and women. Peace and Conflict: Journal of Peace Psychology, 4(4), 381–392. https://doi. org/10.1207/s15327949pac0404_6 Meierrieks, D. (2021). Weather shocks, climate change and human health. World Development, 138(February), 105228. https://doi.org/10.1016/j. worlddev.2020.105228 Meierrieks, D., & Gries, T. (2013). Causality between terrorism and economic growth. Journal of Peace Research, 50(1), 91–104. https://doi. org/10.1177/0022343312445650 Meiyappan, P., & Jain, A. K. (2012). Three distinct global estimates of historical land-cover change and land-use conversions for over 200 years. Frontiers of Earth Science, 6(2), 122–139. https://doi.org/10.1007/s11707-012-0314-2 Montiel Olea, J. L., & Plagborg-Møller, M. (2019). Simultaneous confidence bands: Theory, implementation, and an application to SVARs. Journal of Applied Econometrics, 34(1), 1–17. https://doi.org/10.1002/jae.2656 Mosser, J. F., Gagne-Maynard, W., Rao, P.C., Osgood-Zimmerman, A., Fullman, N., Graetz, N., Roy, B., Updike, R. L., Liu, P.Y., Ray, S. E., Earl, L., Deshpande, A., Casey, D. C., Dwyer-Lindgren, L., Cromwell, E. A., Pigott, D. M., Shearer, F. M., Larson, H. J., Weiss, D. J., & Hay, S. I. (2019). Mapping diphtheria-pertussis-tetanus vaccine coverage in Africa, 2000–2016: A spatial and temporal modelling study. The Lancet, 393(10183), 1843–1855. https://doi.org/10.1016/S0140-6736(19)30226-0 Müller-Crepon, C. (2021). State reach and development in Africa since the 1960s: New data and analysis. Political Science Research and Methods, 11(3), 1–10. https://doi.org/10.1017/psrm.2021.60 Naor, Z. (2006). Untimely death, the value of certain lifetime and macroeconomic dynamics. Defence and Peace Economics, 17(4), 343–359. https:// doi.org/10.1080/10242690600688407 Osgood-Zimmerman, A., Millear, A. I., Stubbs, R. W., Shields, C., Pickering, B. V., Earl, L., Graetz, N., Kinyoki, D. K., Ray, S. E., Bhatt, S., Browne, A. J., Burstein, R., Cameron, E., Casey, D. C., Deshpande, A., Fullman, N., Gething, P.W., Gibson, H. S., Henry, N. J., & Hay, S. I. (2018). Mapping child growth failure in Africa between 2000 and 2015. Nature, 555(7694), 41–47. https://doi.org/10.1038/nature25760 MEIERRIEKS and SCHAUB 40 Peng, J., Simon, D., Hirpa, F., Dyer, E., Lees, T., Miralles, D. G., Vicente-Serrano, S. M., & Funk, C. (2019). High resolution standardized precipitation evapotranspiration index (SPEI) Dataset for Africa. Application/xml. Centre for Environmental Data Analysis (CEDA). https://doi. org/10.5285/BBDFD09A04304158B366777EBA0D2AEB Price, J. I., & Bohara, A. K. (2013). Maternal health care amid political unrest: The effect of armed conflict on antenatal care utilization in Nepal. Health Policy and Planning, 28(3), 309–319. https://doi.org/10.1093/heapol/czs062 Quintana-Domeque, C., & Ródenas-Serrano, P. (2017). The hidden costs of terrorism: The effects on health at birth. Journal of Health Economics, 56(December), 47–60. https://doi.org/10.1016/j.jhealeco.2017.08.006 Reiner, R. C., Graetz, N., Casey, D. C., Troeger, C., Garcia, G. M., Mosser, J. F., Deshpande, A., Swartz, S. J., Ray, S. E., Blacker, B. F., Rao, P.C., Osgood-Zimmerman, A., Burstein, R., Pigott, D. M., Davis, I. M., Letourneau, I. D., Earl, L., Ross, J. M., Khalil, I. A., & Hay, S. I. (2018). Variation in childhood diarrheal morbidity and mortality in Africa, 2000–2015. New England Journal of Medicine, 379(12), 1128–1138. https:// doi.org/10.1056/NEJMoa1716766 Rodríguez, L. (2022). Violence and newborn health: Estimates for Colombia. Health Economics, 31(1), 112–136. https://doi.org/10.1002/hec.4445 Sambanis, N. (2008). Terrorism and civil war. In N. Loayza & P. Keefer (Eds.), Terrorism, economic development, and political Openness (pp.174–206). Cambridge University Press. https://doi.org/10.1017/CBO9780511754388.007 Sandler, T., & Enders, W. (2004). An economic perspective on transnational terrorism. European Journal of Political Economy, 20(2), 301–316. https://doi.org/10.1016/j.ejpoleco.2003.12.007 Schaub, M., & Auer, D. (2023). Rebel recruitment and migration: Theory and evidence from Southern Senegal. Journal of Conflict Resolution, 67(6), 1155–1182. https://doi.org/10.1177/00220027221118258 Schneider, F., Brück, T., & Meierrieks, D. (2015). The economics of counterterrorism: A survey. Journal of Economic Surveys, 29(1), 131–157. https://doi.org/10.1111/joes.12060 Sharara, S. L., & Kanj, S. S. (2014). War and infectious diseases: Challenges of the Syrian civil war. PLoS Pathogens, 10(11), e1004438. https://doi. org/10.1371/journal.ppat.1004438 Sheffield, P.E., & Landrigan, P.J. (2011). Global climate change and children’s health: Threats and strategies for prevention. Environmental Health Perspectives, 119(3), 291–298. https://doi.org/10.1289/ehp.1002233 Sprinzak, E. (1991). The process of delegitimation: Towards a linkage theory of political terrorism. Terrorism and Political Violence, 3(1), 50–68. https://doi.org/10.1080/09546559108427092 Subbarao, K., & Raney, L. (1995). Social gains from female education: A cross-national study. Economic Development and Cultural Change, 44(1), 105–128. https://doi.org/10.1086/452202 Sun, L., & Abraham, S. (2021). Estimating dynamic treatment effects in event studies with heterogeneous treatment effects. Journal of Econometrics, 225(2), 175–199. Themed Issue: Treatment Effect 1. https://doi.org/10.1016/j.jeconom.2020.09.006 Sundberg, R., & Melander, E. (2013). Introducing the UCDP georeferenced event Dataset. Journal of Peace Research, 50(4), 523–532. https://doi. org/10.1177/0022343313484347 Sunstein, C. R. (2003). Terrorism and probability neglect. Journal of Risk and Uncertainty, 26(2), 121–136. https://doi.org/10.1023/A:1024111006336 Tavares, J. (2004). The open society assesses its enemies: Shocks, disasters and terrorist attacks. Journal of Monetary Economics, 51(5), 1039–1070. https://doi.org/10.1016/j.jmoneco.2004.04.009 Tollefsen, A. F., Strand, H., & Buhaug, H. (2012). PRIO-GRID: A unified spatial data structure. Journal of Peace Research, 49(2), 363–374. https:// doi.org/10.1177/0022343311431287 van Kippersluis, H., & Rietveld, C. A. (2018). Beyond plausibly exogenous. The Econometrics Journal, 21(3), 316–331. https://doi.org/10.1111/ ectj.12113 Wagner, Z., Heft-Neal, S., Bhutta, Z. A., Black, R. E., Burke, M., & Bendavid, E. (2018). Armed conflict and child mortality in Africa: A geospatial analysis. The Lancet, 392(10150), 857–865. https://doi.org/10.1016/s0140-6736(18)31437-5 Wagner, Z., Heft-Neal, S., Wise, P.H., Black, R. E., Burke, M., Boerma, T., Bhutta, Z. A., & Bendavid, E. (2019). Women and children living in areas of armed conflict in Africa: A geospatial analysis of mortality and orphanhood. Lancet Global Health, 7(12), e1622–e1631. https://doi. org/10.1016/S2214-109X(19)30407-3 Weiss, D. J., Lucas, T. C. D., Nguyen, M., Nandi, A. K., Bisanzio, D., Battle, K. E., Cameron, E., Twohig, K. A., Pfeffer, D. A., Rozier, J. A., Gibson, H. S., Rao, P.C., Casey, D., Bertozzi-Villa, A., Collins, E. L., Dalrymple, U., Gray, N., Harris, J. R., Howes, R. E., & Gething, P.W. (2019). Mapping the global prevalence, incidence, and mortality of Plasmodium falciparum, 2000–17: A spatial and temporal modelling study. The Lancet, 394(10195), 322–331. https://doi.org/10.1016/S0140-6736(19)31097-9 SUPPORTING INFORMATION Additional supporting information can be found online in the Supporting Information section at the end of this article. How to cite this article: Meierrieks, D., & Schaub, M. (2024). Terrorism and child mortality. Health Economics, 33(1), 21–40. https://doi.org/10.1002/hec.4757