How climate-induced conflict is shaping rural Nigeria

The roots of Nigeria’s herder-farmer conflicts stem, in part, from competition over scarce land and water resources (McGuirk and Nunn 2025). Traditionally, an unwritten rule permits nomadic herders the open use of agricultural land during the dry season, the low period of agricultural activity. Farmers often welcomed the grazing as the wandering cattle naturally fertilised agricultural plots. At the end of the dry season, herders historically vacated agricultural land just in time for farmers to begin cultivating crops.

Yet as climate change extends dry seasons, the migratory patterns of pastoralist herders are being pushed further south, and onto agricultural land, for longer periods of time. Consequently, the annual period preceding the planting season has become a time of increased intensity in violent conflict between farmers and herders.

In response to tensions and outbreaks of violence between herders and farming communities, in 2016 and 2017 three Nigerian states—all traditional herding and farming areas—passed bans on open grazing. The state of Ekiti imposed the first ban in 2016 following the killing of two residents by herders, barring grazing activities in some areas and herders from carrying firearms (any herder found violating the arms ban could be declared a ‘terrorist’ under the statute). In 2017, Benue and Taraba states passed strict open grazing prohibitions. Yet the restrictions did not have the intended result, leading instead to a 2018 spike in violence between herders and farming communities seen in Figure 1.

Figure 1: Violence events in Nigeria, 2004-2019

Notes: Calculations by Bloem et al. (2025) based on ACLED data. Shared with permission by the authors.

It is useful to provide context for the spike in farmer-herder violence both temporally and with reference to other types of violent conflict in Nigeria. From 2010 to 2020, Nigeria experienced a sharp rise in violence propagated by the jihadist group Boko Haram in the northeast region and escalating inter-group conflict between farmers and Fulani pastoralists in the north-central region. Boko Haram’s violent attacks led to states of emergency in both 2011 and 2013; while the group’s activities dominated the news, conflicts between Fulani pastoralists and settled agricultural communities have often been more deadly.

While the absolute number of incidents involving Boko Haram was higher over the 2004-2019 period (see Figure 1), most of those attacks occurred in Borno state, the epicentre of that conflict. By contrast, the herder-farmer conflict grew in both intensity and geographic reach over this period, peaking in 2018 and becoming increasingly more prevalent than violence involving Boko Haram outside of Borno.

How do agricultural households respond to herder-farmer conflict?

Among the many sources of vulnerability smallholder agricultural households face, exposure to violent conflict is increasingly salient. This is especially true in countries with limited capacity to control such conflicts.

This vulnerability is amplified by at least two factors. First, temporal and geospatial patterns of conflict are often closely linked with agricultural seasons (Ubilava 2024), droughts (Merilainen et al. 2023), and contested access to natural resources such as land and water (Dower and Pfutze 2020), thereby placing agricultural households in close proximity. Second, financial markets in low- and middle-income countries, including banks and insurance markets, often function poorly in rural areas and agricultural contexts (Cole et al. 2013, Karlan et al. 2014), and conflict can further hamper their operations.

In Bloem, Damon, Francis, and Mitchell (2025), we study how smallholders respond to these problems, by combing detailed panel data of households and individuals from Nigeria’s General Household Survey (GHS) with data on violent events from the Armed Conflict Location and Event Data (ACLED) project. The GHS data contains four survey rounds, each including two seasonal visits—post-planting and post-harvest—allowing us to observe individuals up to eight separate times between 2010 and 2019.

With this data, we leverage variation across time and space to define our primary indicator of exposure to farmer-herder violence as the presence of farmer-herder violent conflict incidents within a 10-kilometer radius around households taking place within one month before the survey response. To assess previous exposure, we examine violent conflict incidents occurring in the same 10 km area and from one to six months before the survey.

The granularity of this data allows us to include both time and individual fixed effects to estimate changes in economic activities associated with farmer-herder violence while ruling out confounding variation between individuals over time. We also restrict comparisons in two ways: by (i) the planting or harvest seasons and (ii) gender.

Figure 2: Exposure to farmer-herder violence and household labour allocation

Panel A: Household enterprise work

Panel B: Agricultural work

Notes: The vertical axis plots the magnitude of the marginal effect, and associated 95% confidence intervals, of exposure to a herder-related violent event (HRV) on household enterprise work (panel A) and agricultural work (panel B). Sample means of the outcome are in parentheses in the legend. The figure shows results from two specifications: The first does not differentiate by previous exposure. The second differentiates by previous exposure. ‘Singular exposure’ denotes HRV exposure not under the shadow of violence; ‘repeated exposure’ plots the effect of an additional HRV event while also experiencing other recent violence (that is, repeated violence under the shadow of violence). Estimates by Bloem et al. (2025). Shared with permission by the authors.

Farmer-herder violence affects households’ labour allocation decisions

Our findings show that individuals in agricultural households make labour allocation decisions differently based on the timing of conflict exposure relative to three factors: the agricultural season, individual’s gender, and whether exposure represents a singular event or repeated event. The latter finding suggests that these effects seem to operate through a mechanism that we call the ‘shadow of violence’, whereby previous exposure to a violent event casts a shadow over and alters responses to a more recent event.

Our main results, shown in Figure 2, reveal that in the post-planting season, individuals within agricultural households exposed to a single farmer-herder violent conflict event reduce household enterprise work and exhibit no changes in agricultural work. When individuals are exposed to repeated conflict events, however, we observe different household labour responses by gender. During the post-planting season, women work more total hours while men seem to substitute a reduction in agricultural work with an increase in household enterprise work. During the post-harvest season, by contrast, men reduce agricultural work, while women do not.

Furthermore, supplementary analysis shows that exposure to farmer-herder violence does not lead to reported increases in sales from non-farm household enterprises. Thus, while we find that affected households seem to shift labour toward non-farm household enterprises in response to exposure to herder-farmer violent conflict, this behaviour does not seem to effectively guard against adverse economic consequences.

Implications for policy aimed at reducing farmer-herder conflict

Farmer-herder violence can have wide-reaching indirect costs beyond the loss of human life and destruction of property. To address these problems, the ultimate policy goal is to reduce the prevalence of farmer-herder conflict. Yet this is a difficult challenge, as demonstrated by the sharp rise in farmer-herder violence after the implementation of policies that effectively excluded herders and favoured farming communities. McGuirk and Nunn (2025) show, however, that including herding communities in the policy process can moderate farmer-herder violence.

In the meantime, future policy and research efforts could focus on understanding how best to supplement informal coping mechanisms by providing formal economic support for households exposed to violence and conflict. For instance, understanding the gendered nature of household responses to farmer-herder violence motivates interventions that account for the unequal burden of informal coping mechanisms within smallholder agricultural households.

References

Bloem, J R, A Damon, D C Francis, and H Mitchell (2025), “Herder-related violence, labor allocation, and the gendered response of agricultural households,” Journal of Development Economics, 176.

Cole, S, X Gine, J Tobacman, P Topalova, R Townsend, and J Vickery (2013), “Barriers to household risk management: Evidence from India,” American Economic Journal: Applied Economics, 5(1): 104–135.

Dower, P C and T Pfutze (2020), “Land titles and violent conflict in rural Mexico,” Journal of Development Economics, 144.

Karlan, D, R Osei, I Osei-Akoto, and C Udry (2014), “Agricultural decisions after relaxing credit and risk constraints,” Quarterly Journal of Economics, 129(2): 597–652.

McGuirk, E F and N Nunn (2025), “Transhumant pastoralism, climate change, and conflict in Africa,” Review of Economic Studies, 92(1): 404–441.

Merilainen, J, M Mitrunen, and T Virkola (2023), “Famine, inequality, and conflict,” Journal of the European Economic Association, 21(4): 1478–1509.

Ubilava, D (2024), “Climate, crops, and post-harvest conflict,” American Journal of Agricultural Economics.