The impact of climate change on the emergence of violent conflicts has become the subject of lively debate in recent years. However, despite the substantial number of studies, scholars have as yet been unable to come up with a general interpretation of the climate change–violence nexus. In fact, the results so far are puzzled, contradictory and inconclusive.
Several aspects contribute to the inconclusiveness of the literature. First, the causal mechanism behind the climate change–violence relationship may vary across societies, regions and territories. In the eyes of economists, phenomena of climate change result in exogenous shocks that significantly affect agricultural production, thus shaping individuals’ economic opportunities, and thereby their incentives and opportunity costs.
Second, among social scientists, climate change has been frequently approximated in the literature on conflict by means of rainfall and mean temperature. However, the natural science literature claims that minimum temperatures have a strong impact on cereal production. Higher minimum temperatures increase the maintenance respiration requirements of both rice and wheat, and shorten the time to maturity, thus reducing net growth and productivity.
Since the minimum temperature is usually reached during the night, by smoothing the variations across the 24-hour temperature range, scholars might have failed to consider the actual extent of the connection between climate change and conflict. In particular, a substantialbody of evidence claims that climate change in Indonesia has a negative effect on rice production.
Starting from this evidence, we analysed the link between climate change and conflict in 14 Indonesian provinces in the period 1990–2003. We chose Indonesia for two reasons. First, it is a top rice producer and has been affected by several waves of violence over the years. Indeed, Indonesia has a well-studied history of violence whose escalation in the late 1990s was accompanied by three transitions: (a) a political transition from autocracy to democracy, (b) an economic transition from crony capitalism to ruled-based market systems; and (c) a social transition from state centralization to decentralization.
Second, the overall severity of conflict in Indonesia peaked after the fall of the Suharto regime in 1998 and then decreased after 2001. The reduction in violence can be explained in the light of fiscal decentralization (Murshed and Tadjoeddin, 2008), repression of terrorist networks in the aftermath of the Bali bombing in 2002, and the end of the long-lasting separatist war in Aceh.
However, despite the decreasing pattern of conflict severity, the number of violent incidents rose again after 2002. Moreover, according to the Food and Agriculture Organization dataset, rice is the main crop in Indonesia in terms of both quantity and value as well as the top commodity available for internal consumption. In this respect, it is necessary to note that despite being a top rice producer, Indonesia has historically been a net importer of rice.
We study two indicators of violence, namely the emergence of violence measured by number of monthly events, and the severity of violence measured by means of the number of monthly victims (sum of injured and killed people). Using an econometric approach that allows us to claim that our results are causal and not just correlations, we show that an increase of the minimum temperature during the core month of the ‘wet planting season’ (December), determines an increase in violence fuelled by the reduction in future rice production.
Therefore, we find support for the hypothesis that minimum temperature negatively affects rice availability (per capita), which in turn inflames violence. This evidence remains robust under alternative specifications of the dependent variable, as well as the inclusion of a number of controls.
Stated broadly, our work suggests that some inconclusive results in the literature on the climate change–violence nexus could suffer from failing to consider the specificities of different crops.
The economic relevance and type of the dominant agricultural output—be it rice, wheat or another crop—changes across countries. Crops are heterogeneous in terms of growing season, cultivation technology and reaction to climate variability. Further research seems necessary in order to explore other combinations of climate change, crops and violence in different countries.
The impact of climate change on the emergence of violent conflicts has become the subject of lively debate in recent years. However, despite the substantial number of studies, scholars have as yet been unable to come up with a general interpretation of the climate change–violence nexus. In fact, the results so far are puzzled, contradictory and inconclusive.
Several aspects contribute to the inconclusiveness of the literature. First, the causal mechanism behind the climate change–violence relationship may vary across societies, regions and territories. In the eyes of economists, phenomena of climate change result in exogenous shocks that significantly affect agricultural production, thus shaping individuals’ economic opportunities, and thereby their incentives and opportunity costs.
Second, among social scientists, climate change has been frequently approximated in the literature on conflict by means of rainfall and mean temperature. However, the natural science literature claims that minimum temperatures have a strong impact on cereal production. Higher minimum temperatures increase the maintenance respiration requirements of both rice and wheat, and shorten the time to maturity, thus reducing net growth and productivity.
Since the minimum temperature is usually reached during the night, by smoothing the variations across the 24-hour temperature range, scholars might have failed to consider the actual extent of the connection between climate change and conflict. In particular, a substantial body of evidence claims that climate change in Indonesia has a negative effect on rice production.
Starting from this evidence, we analysed the link between climate change and conflict in 14 Indonesian provinces in the period 1990–2003. We chose Indonesia for two reasons. First, it is a top rice producer and has been affected by several waves of violence over the years. Indeed, Indonesia has a well-studied history of violence whose escalation in the late 1990s was accompanied by three transitions: (a) a political transition from autocracy to democracy, (b) an economic transition from crony capitalism to ruled-based market systems; and (c) a social transition from state centralization to decentralization.
Second, the overall severity of conflict in Indonesia peaked after the fall of the Suharto regime in 1998 and then decreased after 2001. The reduction in violence can be explained in the light of fiscal decentralization (Murshed and Tadjoeddin, 2008), repression of terrorist networks in the aftermath of the Bali bombing in 2002, and the end of the long-lasting separatist war in Aceh.
However, despite the decreasing pattern of conflict severity, the number of violent incidents rose again after 2002. Moreover, according to the Food and Agriculture Organization dataset, rice is the main crop in Indonesia in terms of both quantity and value as well as the top commodity available for internal consumption. In this respect, it is necessary to note that despite being a top rice producer, Indonesia has historically been a net importer of rice.
We study two indicators of violence, namely the emergence of violence measured by number of monthly events, and the severity of violence measured by means of the number of monthly victims (sum of injured and killed people). Using an econometric approach that allows us to claim that our results are causal and not just correlations, we show that an increase of the minimum temperature during the core month of the ‘wet planting season’ (December), determines an increase in violence fuelled by the reduction in future rice production.
Therefore, we find support for the hypothesis that minimum temperature negatively affects rice availability (per capita), which in turn inflames violence. This evidence remains robust under alternative specifications of the dependent variable, as well as the inclusion of a number of controls.
Stated broadly, our work suggests that some inconclusive results in the literature on the climate change–violence nexus could suffer from failing to consider the specificities of different crops.
The economic relevance and type of the dominant agricultural output—be it rice, wheat or another crop—changes across countries. Crops are heterogeneous in terms of growing season, cultivation technology and reaction to climate variability. Further research seems necessary in order to explore other combinations of climate change, crops and violence in different countries.
This blog post is published as part of a collaborative partnership between SIPRI and Economists for Peace and Security (EPS).
ABOUT THE AUTHOR(S)