Environment, Development and Sustainability

https://doi.org/10.1007/s10668-024-05358-y

Climate change and political stability in African states:

the role of agricultural food price volatility

Ateba Boyomo Henri Aurélien1 · Ongo Nkoa Bruno Emmanuel2

Received: 26 October 2022 / Accepted: 25 August 2024

© The Author(s), under exclusive licence to Springer Nature B.V. 2024

Abstract
In a context marked by the increase in extreme climatic shocks, and the decline in the polit-
ical stability of States in Sub-Saharan Africa (SSA), this article aims to analyze the effect
of climate change on the political stability of 45 countries of sub-Saharan Africa over the
period 2000–2020. Thus, we use time-country fixed effects to mitigate possible endogene-
ity biases. Our results obtained by Two-Way Fixed effets (TWFE) estimator indicate that
climate change reduces political stability in SSA. Indeed, rising temperatures and falling
precipitation reduce the political stability of these countries and vice versa. Furthermore,
through the marginal effects, we show that wheat price volatility reduces the significant and
positive effect of heavy rainfall on political stability, while rising wheat prices aggravate
the negative effect of rising temperatures on political stability. The Generalised Method of
Moments in System (GMM-S) and the method of lagged explanatory variables allowed us
to better take into account the endogeneity problem, and to validate the robustness of our
results. To reduce the risk of political instability in SSA, we recommend the development
of effective methods of adaptation or mitigation of the effects of climate change, through
the preservation of the environment, the promotion of green energies and the strengthening
of infrastructure water management and storage, for agricultural populations. Furthermore,
stabilizing food prices requires increasing the level of national and regional stocks of vari-
ous foodstuffs.

Keywords Political stability · Climate change · Food price volatility · Africa

JEL Classification Q11 · Q51 · Q54 · C23

* Ateba Boyomo Henri Aurélien

[email protected]
Ongo Nkoa Bruno Emmanuel
[email protected]
1
Center for Studies and Research in Management and Economics, Faculty of Economics
and Management (FSEG), University of Yaoundé II, PO Box 1365, Yaoundé, Cameroon
2
Center for Studies and Research in Management and Economics, Faculty of Economics
and Management (FSEG), University of Yaoundé II, PO Box 1365, Yaoundé, Cameroun

Vol.:(0123456789)
 A. B. Henri Aurélien, O. N. Bruno Emmanuel

1 Introduction

Political stability is defined as the measure of perceptions about the likelihood of the gov-
ernment being destabilised or overthrown by unconstitutional means, including domestic
violence and terrorism (Radu, 2015). Therefore, an unstable political environment will
increase the probability of government collapse and political unrest (Alesina & Perotti,
1996). Several authors consider social unrest, anti-government protests, political violence,
political assassinations and government stability as measures of political instability (But-
kiewicz & Yanikkaya, 2006). For Gupta (1991), political instability can take three forms,
namely elitist or executive instability coups d’état, changes or crises of government, mass
instability referring to social movements such as strikes, demonstrations or riots, and
armed or violent instability including civil wars and guerrilla wars.
In the context of Africa in general, it is the third form of instability that is most
observed. Indeed, based on the index constructed by Maplecroft, 5 African countries were
classified in 2014 as being at extreme risk of political violence, and 10 are at high risk.
Moreover, according to the political risk map, 14 of the main hotspots of political tension
are in Africa. The empirical literature identifies several determinants of political instability,
such as the level of growth (Alesina & Perotti, 1996; Collier & Hoeffler, 2004), the level of
education (Xu, 2011), and the annual growth rate of the urban population (Manarik, 1981).
Other studies identify natural resources (Dupasquier & Osakwe, 2006), migrant remit-
tances (Avom et al., 2020), corruption (Eggoh et al., 2021), and foreign direct investment
(Gakpa, 2019) as sources of political instability. From the point of view of theories, the
Malthusian approach and political economy make it possible to highlight the effect of cli-
mate change on political instability. The Malthusian approach to the scarcity of resources
developed by Malthus (1798) raises the problem of the resurgence of potential conflicts
following the scarcity of resources. This theory has been taken up to explain the emergence
of conflicts related to environmental resources by the schools of Thomas Homer-Dixon
(1999) and his colleagues at the University of Toronto on the one hand, and that of the
Swiss Bächler and Spillman (1996) of Environment and Conflicts Project (ENCOP) on the
other hand. These two schools conclude that the degradation and reduction of agricultural
land, forests, water resources, and even fish stocks contribute to the appearance of armed
conflicts.
Moreover, the foundations of political stability are based on the theory of the politi-
cal economy of armed conflict developed over the last three decades, which considers
armed conflicts as structures of political and economic opportunities (Hugon, 2009). This
includes the theory of greed (Collier & Hoeffler, 2004), the theory of natural resource pre-
dation (Bannon & Collier, 2003), and the grievance theory (Berdal & Malone, 2000).
However, although there is no evidence of a long-term increase in food price volatility,
there is no doubt that the beginning of 2006 was a period of extraordinary volatility affect-
ing households and businesses (FAO, 2011). For this reason, the food crisis of 2007–2008,
which was caused by rising food prices, resulted in protests that posed a risk of instability
for developing countries and their households. There are two reasons why this price vola-
tility poses a risk to these African households. On the one hand, it tends to be higher in
Africa than in other regions (Minot, 2014), on the other hand, these households spend 60%
of their budget on food, so an increase in international prices could reduce their purchasing
power and leat to protests (FAO, 2011).
Furthermore, the threat multiplier approach is to our knowledge one of the main theoret-
ical approaches highlighting the role of climate change on the political instability of states.
Climate change and political stability in African states: the…

Indeed, the UN Security Council (2019) recognizes that climate change can act as a ’threat
multiplier’, aggravating certain stresses that are often at the heart of state instability. For
example, according to a report by Women UN (2022), climate change intensifies social,
economic, political and security tensions.
According to the empirical literature, several studies highlight the effect of climate
change on conflicts, which may vary from one conception to another. These include inter-
personal conflicts (Koubi, 2019; Mares & Moffett, 2016), intergroup conflicts between
states (Devlin & Hendrix, 2014), or against the government (Ateba et al., 2023; Burke
et al., 2009), civil war (Von Uexkull et al., 2016; Raleight et al., 2015; Buhaug et al., 2015;
Demarest (2014); Berazneva & Lee, 2013), communal violence (Detges, 2016), and polit-
ical repression (Wood & Wright, 2016). Moreover, among the existing studies, those of
Raleigh and Urdal (2007), Raleigh and Kniveton (2012), and Koubi (2019) recognize a
direct effect of climate change on conflict and political tension, while others highlight an
indirect influence of change on conflict (Berazneva & Lee, 2013; Demarest, 2015; Koubi
et al., 2012). Indeed, in terms of the indirect effect, long-term food supply concerns are
driven by the fact that climate change is expected to increase the prices of food and agricul-
tural commodities such as rice, wheat, maize, and soy (Nelson et al., 2009), thus increas-
ing conflicts. In addition, other analyses show that the number of people at risk of hun-
ger and conflict is projected to increase by 10–20% by 2050 due to climate change, with
approximately 65% of this increase occurring in Africa (Parry, 2009). However, apart from
the work of Sofuoğlu and Ay (2020) highlighting the existence of a causal relationship
between climate change and political instability in the countries of the Middle East and
North Africa (MENA), to our knowledge, there is no work on the issue in Sub-Saharan
Africa. Indeed, the work of Eggoh & Kobbi (2015) distinguishes informal political insta-
bility (coup d’état, internal and external conflicts, intergroup and interpersonal conflicts,
assassination, strikes, and political tensions), from formal political stability (crisis at the
top of the state following the constitutional and legislative change). Moreover, unlike the
work of Sofuoğlu and Ay (2020) which measures political instability by the political insta-
bility index, we measure it by the political stability index which contains more advantages.
Specifically, our study has five contributions: (i) it enriches the literature and fills the gap
in studies on the effect of climate change on political instability in SSA; (ii) In addition to
other studies that are limited to the effect of climate change on the informal dimension of
political instability (conflicts), our study introduces the formal dimension of political insta-
bility through the use of the political stability index; (iii) unlike Sofuoğlu and Ay (2020)
which use the Political Instability Index from the International Country Risk Guide (2018)
to capture political instability, we use the Political Stability Index from the World Govern-
ance Indicator (WGI, 2020), which has the advantage of providing information for a large
number of countries at the same time, and over a longer recent time horizon. (iv) we use
in this paper a time and country fixed effects model based on the TWFE estimator, which
is more appropriate in the literature when dealing with the climate-conflict relationship,
to address potential endogeneity biases (Hsiang et al., 2013; Weezel, 2020). (v) Finally,
this work does not take into account the volatility of food prices as a transmission channel
between climate change and political instability, even though it is recognized that the rising
prices of agricultural commodities such as wheat (which is climate-dependent) are at the
root of riots and protests. The remainder of the paper is organized as follows: in Sect. 2,
we provide some stylized facts on the subject, followed by a review of the literature on the
issue in Sect. 3. In Sect. 4, we present the methodology of the study, followed in Sect. 5
by the presentation of the results of the econometric estimations. Section 6 of the paper
proposes a discussion of the results through robustness tests and some recommendations.
 A. B. Henri Aurélien, O. N. Bruno Emmanuel

2 Review of the literature

This section presents a review of the literature on the impact of climate change on the
resurgence of political instability in Africa.

2.1 Effect of climate change on political stability: theorical bachground

In this study, we mobilize the Malthusian theory of resource scarcity and the theory of
political economy as theoretical foundations of the effect of climate change on political
instability.

2.1.1 Malthusian theory of resource scarcity and its extensions

Developed in 1978 by Malthus, the theory of resource scarcity states that if the population
increases in geometric progression, subsistence increases only in arithmetic progression.
This assertion by Malthus (1978) poses the problem of the resurgence of potential conflicts
between individuals following the scarcity of resources. This theory has been taken up to
explain the emergence of conflicts related to environmental resources by the schools of
Homer-Dixon (1999) and his colleagues at the University of Toronto on the one hand, and
that of the Swiss Bächler and Spillman of the ENCOP on the other hand. these two schools
conclude that the degradation and reduction of agricultural land, forests, water resources,
and even fish stocks contribute to the appearance of armed conflicts. According to Le Ster
(2011), the thesis of conflicts linked to the scarcity of environmental resources as devel-
oped by these schools contains parallels with market economic theories. We find there the
environmental scarcity induced by the supply (the probability of the emergence of vio-
lent conflicts would be greater in the States where there is land degradation, deforestation,
or where the water resources available per capita are low), demand-driven environmental
scarcity (the likelihood of the existence of armed conflicts is greater in countries facing
high population density), and structural environmental scarcity (states with high-income
inequality seem more exposed to the risk of conflict).

2.1.2 The theory of the political economy of conflict

The political economy approach to the origins of armed conflict developed over the last
three decades has long considered conflicts as a structure of political and economic oppor-
tunities (Hugon, 2009). It also includes: (i) the theory of greed, whose economic analysis
considers rebellion not as a means of vindication and social justice, but rather as a form of
organized crime.
This theory attributes economic behavior to social actors in situations of armed con-
flict. Thus, these actors are guided by a logic of economic opportunity (Collier & Hoef-
fler, 2004). (ii) The theory of natural resource predation, developed by Bannon and Collier
(2003), demonstrates that the control of natural resources is a central issue in wars. Thus, a
country’s dependence on its natural resources can foster situations of armed conflict. These
analyses are supported by the work of Ballentine and Nitzschke (2003) who have dem-
onstrated the undeniable character of natural resources in the development of civil war.
Finally, (iii) the grievance theory which states that armed conflicts emanate from several
Climate change and political stability in African states: the…

causes, in particular social inequalities and distortions, long-lasting socio-economic injus-

tices, and ethnic community divisions (Berdal & Malone, 2000).

2.2 Empirical literature: direct effect of climate change and political stability

Very few studies highlight the effect of climate change on political stability. Indeed, the
most recent work is that of Sofuoğlu and Ay (2020), which in the case of Middle East-
ern and North African countries, shows that there is a causal relationship between climate
change and political instability. Therefore, the empirical results support the hypothesis that
climate change acts as a threat multiplier in MENA countries as it triggers, accelerates, and
deepens current instabilities.
However, there is an extensive literature on the direct effect of climate change on con-
flict. According to the Intergovernmental Panel on Climate Change (IPCC, 2014), there
is a clear link and direct between climate and conflict. According to Koubi (2019), two
explanations make it possible to validate the direct effect of climate change on conflicts:
psychological and/or physiological factors, and factors related to the scarcity of avail-
able resources. Concerning psychological and/or physiological factors, analyses show that
hotter and colder temperatures, by increasing the level of discomfort and aggressiveness
of individuals, increase interpersonal conflicts (Anderson & Bushman, 2002). Moreo-
ver, based on the hypothesis of the scarcity of resources included in the neo-Malthusian
approach, Koubi (2019) tells us that the scarcity of water resources due to extreme tem-
peratures and the drop in the precipitation regime generates conflicts between the popula-
tions that depend on them. Thus, a decrease in resources due to reduced precipitation in a
region increases the risk of conflict (Raleigh & Urdal, 2007). In addition, Homer-Dixon
(1994) demonstrates that the scarcity of natural resources due to climate change generates
national and international conflict, a result confirmed by the work of Hendrix and Salehyan
(2012), and Theisen (2012). Jensen and Gleditsch (2009) show that decreasing rainfall
increases the outbreak of civil wars. Other studies further show that low or declining rain-
fall increases the risk of communal conflict, such as Hindu-Muslim riots in India (Bohlken
& Sergenti, 2010; Sarsons, 2011).
Nevertheless, certains studies show that high temperatures increase the risk of many
forms of intergroup and political violence (Burke et al., 2009; Caruso et al., 2014; Dell
et al., 2012; Maystadt & Ecker, 2014; Maystadt et al., 2015; O’Loughlin et al., 2012), oth-
ers conclude that high temperatures and excessive precipitation variability do not influence
the risk of armed conflict (Buhaug, 2010; Theisen, 2012).
In the African context, there is strong and consistent evidence that higher temperatures
lead to increased civil conflict (Hsiang et al., 2013).
In addition, Von Uexkull (2014) finds that regions in sub-Saharan Africa that are par-
ticularly dependent on rainfall for agricultural production are also more likely to experience
civil conflict. In the case of East Africa, Raleigh and Kniveton (2012) show that higher
rainfall results in less conflict, while drier conditions increase conflict. From the above, we
formulate hypothesis 1 that: climate change has a direct effect on political stability in SSA.

2.3 Food price volatility as transmission channel little used in the literature

Some studies mention the fact that food prices are observed as intermediate variables, but
rarely used to present the link between climate change and conflict (see Berazneva & Lee,
2013; Zhang et al., 2010; Koubi et al., 2012). Existing studies show that the increase in
 A. B. Henri Aurélien, O. N. Bruno Emmanuel

polical stability index

0

2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
-0.1

-0.2

-0.3

-0.4

-0.5

-0.6

-0.7

-0.8
Axis Title

Fig. 1  Evolution of political stability in sub-Saharan Africa between 2000 and 2020. Source Author based
on data from the Worldwide Governance Indicator (WGI, 2020)

agricultural commodity prices due to climate disruption creates shortages and compe-
tition for natural resources, which generates conflicts (Demarest, 2015). Climate change
as a source of conflict in Africa through soaring food prices has raised the issue of food
riots (Bellemare, 2014; Hendrix et al., 2009; Lagi et al., 2011). Furthermore, Raleigh et al.
(2015) demonstrates that climate change influences conflict through rising food prices.
Moreover, the scarce literature on the subject is limited by the fact that it deals with the
indirect effects of climate change on conflicts, without mentioning political stability, which
is our concept of analysis.

3 Some stylized facts

Two stylized facts can emerge from this study. The first shows the evolution of politi-
cal stability and climate change, while the second analyses the correlation between these
variables.

3.1 Political stability in SSA declines significantly

We highlight in Fig. 1 below the evolution of political stability and climate change in SSA.
Firstly, this graph shows that sub-Saharan Africa is generally politically unstable.
Indeed, between 2000 and 2007, the political stability index1 reflects instability, since its
value is between −0.54 and −0.53. However, from 2008 onwards, which was characterized
by food riots, there was an increase in political instability, with the value rising to −0.58.
Furthermore, between 2014 and 2020, there was a significant increase in political instabil-
ity, with an average value of −0.72, explained in part by the formation of numerous jihadist
groups in the Sahelian regions, and the multiple post-electoral crises.

1
We recall that the World Bank’s Worldwide Governance Indicator ranges from −2.5 (for political
instability) to 2.5 (for political stability).
Climate change and political stability in African states: the…

28.1 1020

Average precipitaon mm/Year

28 1000
Average temperature (°C)

27.9 980
27.8
960
27.7
940
27.6
920
27.5
27.4 900
27.3 880
27.2 860
27.1 840
27 820
2000 2005 2010 2015 2020 2000 2005 2010 2015 2020
Year Year

Fig. 2  Temperature (°C) and precipitation (mm/year) trends in SSA. Source Authors using data from
Climate Change Knowledge Portal (CCKP, 2021)

3.2 Climate change persists in SSA

In addition, although Fig. 2 shows an abundance of rainfall between 2001 and 2005, it also
shows that rainfall decreased and became erratic during the rest of the period. Furthermore,
although the United Nations report (UN, 2020) indicates that the decline in rainfall patterns
in SSA is estimated to be 20% by 2020, Fig. 2 shows that the trend is a slight decrease in
precipitation. The recurrent fluctuations observed in rainfall reflect climate change, i.e.
a change in the structure of rainfall, which is no longer seasonal but irregular. Overall,
these analyses of the persistence of climate change in SSA are supported by the World
Meteorological Organisation (WMO, 2020) report on climate change. Indeed, the WMO
(2020) indicates that the African trend is towards warming, with temperature levels whose
average variation will be around + 0.3 °C in 2022, as opposed to the average variation
of + 0.2 °C observed in the rest of the world. The continent is also experiencing rainfall
anomalies, the main characteristics of which are rainfall deficits and irregularities in the
occurrence of the first rainfall of the year. The most affected regions are West Africa, the
Eastern Sahel, the Horn of Africa and parts of Southern Africa (WMO, 2020).

3.3 Climate change and political stability are correlated

Figures 3 and 4 show that climate change and political stability are correlated. Indeed, Fig. 3
shows an inverse relationship between increasing temperatures and political stability, since
any increase in temperature decreases the political stability of states. Similarly in Fig. 4,
decreasing rainfall decreases the political stability of states and vice versa, so that favorable
rainfall could increases political stability. These analyses are in line with the findings of
the IPCC (2014) which recognize an unmistakable effect of rising temperatures and falling
rainfall on instability (Fig. 5).
 A. B. Henri Aurélien, O. N. Bruno Emmanuel

Fig. 3  Correlation between variation of average temperature and political stability. Source Author based on
World Bank Group’s Climate Change Knowledge and WGI (2020)

Fig. 4  Correlation between variation of average precipitation and political stability. Source Author based
on World Bank Group’s Climate Change Knowledge and WGI (2020)

4 Empirical strategy

This section aims to provide a methodological analysis of our study. It will present the
analysis model and data, and the estimation technique.

4.1 Model and data

This article proposes to estimate a time-country fixed effects panel model contained in the
following Eq. (1), to minimize statistical bias:
Climate change and political stability in African states: the…

Climate Change Political Stability

Mediator

Variation of average temperature Agicultural Food price volatility Political stability index
Variation of average precipitation (Wheat)

Fig. 5  Highlights the causality diagram between the variables in the study. Source Author’s construction

PSIit = 𝛽0 + 𝛽1 CCit + 𝛽2 Xit + 𝜓i + 𝜌t + 𝜀it (1)

i = 1, 2…0.45 and t = 2000,2,001….,2020.
Indeed, the results of the Hausman2 test reject the null hypothesis of the absence of ran-
dom effect in our sample, thus favoring the use of the fixed-effect model.
Where ρt and Ψi correspond to the temporal and individual dimensions, respectively.
Furthermore, to analyze the effect of climate change and food price volatility on politi-
cal stability, we used several databases over the period 2000–2020, for a panel of 45 sub-
Saharan African countries. According to these models, the introduction of the individual
and temporal dimension makes it possible to mitigate the potential biases of endogene-
ity, in particular of unobserved heterogeneity and of correlation between the model vari-
ables. Concretely, these models differ from cross-sectional models which, in addition to
not taking into account the specific fixed effects of the level of conflict and climate for each
individual, do not measure the severity of conflict and climate variability between these
countries (Hsiang, Burke and Miguel, 2014). ­PSIit is the political stability index and meas-
ures the political stability of states. The Political Stability and Freedom from Violence/
Terrorism Index measures the perceived likelihood of the government being destabilized or
overthrown by unconstitutional or violent means, including politically motivated violence
and terrorism (Radu, 2015).
Unlike the International Country Risk Guide (ICRG) database, this database provides
governance information for a larger number of countries, and over a longer time horizon.
However, the World Bank report (2013) on the construction of the index of political stabil-
ity and the absence of violence indicates that it is constructed from several variables from
different sources, the most important of which for us is the stability of the government
of the Political Risk Service (PRS). Thus, for Yabré et al. (2020), the work of Alesina
(1989) political stability to institutional stability of governments measured by government
unity and legislative support. For robustness, we will use the rule of law index to meas-
ure the political stability of states. The rule of law captures perceptions of the extent to
which agents trust and abide by the rules of society, and in particular the quality of contract
enforcement, property rights, police and courts, and the likelihood of crime and violence
(Radu, 2015). In the end, the advantage of directly using indices rather than disaggregated

2
chi2(13) = (b-B)’[(V_b-V_B)^(-1)](b-B) = 20.75; Prob > chi2 = 0.0780.
 A. B. Henri Aurélien, O. N. Bruno Emmanuel

values helps to solve the problems of multicollinearity (Keho, 2012). ­VOFPit represents
the volatility of food prices, and is derived from FAO monthly data (Faostat, 2020). Hugon
and Mayeyenda (2003) use the coefficient of variation to measure price volatility. But for
our study, based on the work of Minot (2014), we we obtain the volatility of food prices,
by calculating the standard deviations3 on the monthly food consumer price index data.
Indeed, the authors show that as opposed to the coefficient of variation, the deviation is
more appropriate for measuring volatility. We will focus on the volatility of the price of
wheat since it is the food that is at the origin of tensions and strikes (Berazneva & Lee,
2013).
CCit is the matrix of climate change variables from the World Bank Group’s Climate
Change Knowledge Portal (CCKP, 2021) monthly databases. It includes the variation in
average temperature ­(VATit) and the variation in average precipitation ­(VAPit) obtained as
in Ateba et al. (2024), by calculating annual standard deviations from monthly temperature
and precipitation.
To test the quality of our results, we introduce X ­ it the matrix of control variables. It
includes GDPit, represents gross domestic rate per capita, and measures the level of eco-
nomic activity and corresponds to the ratio between the value of GDP and the number of
inhabitants of a country. As in Fosu (2003), economic growth increases political stability.
COCit represents the Control of corruption variables which captures perceptions of the
extent to which public power is exercised for private gain, including both petty and grand
forms of corruption, as well as the “capture” of the state by elites and private interests.
For Eggoh et al. (2021), controlling corruption reduces the political instability of states.
Educit measures the number of students enrolled in secondary school. For Xu (2011)
education increases political stability. ­Popgrit measures the rate of population growth based
on the de facto definition of population, which counts all residents regardless of legal status
or citizenship—except for refugees not permanently settled in the country of asylum, who
are generally considered part of the population of the country of origin. Manarik (1981)
shows that, a high population expansion can be the cause of population uprisings. N ­ atresit
is natural resources, approximated by profits from oil, gas, and minerals as a percentage
of GDP. According to the natural resource curse, a country with an abundance of natural
resources is more prone to conflict (Dupasquier & Osakwe, 2006).
Remfit represents migrant remittances and corresponds to the share of income earned
abroad. As demonstrated by Avom et al. (2020), remittances reduce the political stability of
already fragile states. ­FDIit is the net inflow of foreign direct investment (as a % of GDP).
For Gakpa (2019) FDI and political instability are highly correlated. All these data are taken
from the World Bank’s World Development Index (WDI, 2020), the Food and Agriculture
Organization (FAOSTAT, 2020) and the World Governace Indicator (WGI, 2020). The table
in Annex 2 provides sufficient information on the variables and their different sources.
From the above, the model in the extended form is written:
PSIit = 𝛽 + 𝛾1 VATit + 𝛾2 VAPit + 𝜆1 GDPit +𝜆2 POPGRit + 𝜆3 NATRESit + 𝜆4 REMFit
+ 𝜆5 FDIit +𝜆6 EDUCit +𝜆7 COCit +𝜆8 VOFPit + 𝜓i + 𝜌t + 𝜀it
(2)
Finally, we recall that we introduce in the model an interaction variable between food
price volatility and climate change noted VOFP*CCit. According to Raleigh et al. (2015),

3
As in Minot (2014), this involves the calculation of the standard deviation of the yield noted σ (r) with
r = log (Pt / Pt − 1), and Pt and Pt − 1, the prices at period t and t-1.
Climate change and political stability in African states: the…

price volatility is a transmission channel between climate change and political stability. It
is an observed variable, but very rarely used in the literature (Berazneva & Lee, 2013) and
Koubi et al., 2012). Thus, our model is written:
PSIit = 𝛽0 + 𝛽1 CCit + 𝛽2 Xit + 𝛽3 VOFPit ∗ CCit + 𝜓i + 𝜌t + 𝜀it (3)
From the above, our model taking into account the interactions between climate change
and the volatility of food prices in the extended form is written as:
PSIit = 𝛽 + 𝛾1 VATit + 𝛾2 VAPit + 𝜆1 GDPit +𝜆2 POPGRit + 𝜆3 NATRESit + 𝜆4 REMFit + 𝜆5 FDIit +𝜆6 EDUCit
+𝜆7 COCit +𝜆8 VOFPit + 𝛾1 VATit ∗ VOFPit + 𝛾2 VAPit ∗ VOFPit 𝜓i + 𝜌t + 𝜀it
(4)
The analyses of the descriptors on the mean and correlation are presented in Appen-
dix 3 and 4. It can be seen that all the variables are clustered around their respective means
and that the standard deviations (Appendix 3) are very low in volatility. Furthermore, the
analysis of the correlation (Appendix 4) indicates a very low correlation between the vari-
ables, thus indicating a presumption of the absence of multicollinearity. The results of the
multicollinearity test of the variance of inflation factor (VIF) contained in Appendix 5,
give for all the explanatory variables of the model coefficients lower than 10, which indi-
cates our model does not suffer from multicollinearity. In addition, the Breush–Godfrey
(Appendix 6) and Breush-Pagan (Appendix 7) tests validate, respectively, in our sample,
the absence of autocorrelation of errors and heteroscedasticity. Below is the causal diagram
linking our variables (Fig. 5).

4.2 Estimation method

In this study, we apply the two-way fixed effects (TWFE) method to estimate our model.
Indeed, thanks to the introduction of the time and individual dimension, this estimation
technique mitigates the problem of endogeneity bias (Burke et al., 2009, 2015; Eberle
et al., 2020; Hsiang et al., 2013; Koubi, 2019). Specifically, this technique is used to atten-
uate unobserved heterogeneity between individuals in the panel (Imai & Kim, 2021; Wool-
dridge, 2021). In our study, we show that: (i) thanks to the introduction of specific coun-
try fixed effects, this method controls for unobserved heterogeneity between countries in
the panel, which may be related to cultural, historical, political, economic, geographical
or institutional differences for all countries in sub-Saharan Africa. (ii) The introduction of
the time dimension allows the estimator to mitigate the problem of endogeneity due to a
suspected correlation between climate change, political stability and control variables such
as population growth rate and GDP per capita. Finally, the time dimension captures all the
shocks common to the countries in my panel.
The choice of this technique is validated by the Fisher test which rejects the null hypoth-
esis of homogeneity of individual dimensions.4 However, as a robustness estimator, we use
the GMM-S5 which, like the TWFE, takes into account the problems of endogeneity. more-
over, thanks to the non-parametric method focused on the estimation (by bidirectional fixed
effects) of a variable model of explanatory variables lagged by one or two periods, we take

4
F( 20, 896) = 8.48 with Prob (F-statistic) = 0.0000.
5
The Generalized Method of Moments (GMM), whose Generalized Method of Moments in Difference
(GMM-D) estimator was introduced by Arellano and Bond (1991). Subsequently, Blundell and Bond
(1998) refined the GMM-D estimator by developing the Generalized Method of Moments in System
(GMM-S) estimator.
 A. B. Henri Aurélien, O. N. Bruno Emmanuel

more into account questions of endogeneity, in particular those of simultaneity, multicol-

linearity and overparameterization (Bellemare et al., 2017).

5 Baseline results

This section presents our main results on the direct and indirect effects of climate change
on political stability.

5.1 Direct effects

The results in Table 1 show that for almost all specifications (1–10), climate change influ-
ences the political stability of states. Indeed, the results show that there is an inverse rela-
tionship between the variation in average temperatures and the political stability of States,
such that an increase in temperatures leads to a decrease in the political stability of States,
and therefore an increase in political instability. In this sense, as in Raleigh and Kniveton
(2012) and Raleigh et al. (2015), any increase in temperature leads to an increase in politi-
cal instability. Likewise, variation in average rainfall positively influences the political sta-
bility of states, such that abundant rainfall stabilizes African states with agricultural GDP.
And conversely, any reduction in precipitation generates a reduction in political stability in
SSA. This is in line with the work of Raleigh and Urdal (2007) and Lecoutere et al. (2010),
for whom the reduction of precipitation generates conflicts. This result of the direct effect
of climate change on political stability in SSA is justified by the neo-Malthusian approach
to resource scarcity (Koubi, 2019). To the extent that any reduction in natural resources
such as water, land and forests created rivalry, and therefore conflicts between the popu-
lations who spend them. As for the effect of control variables, we show for all specifica-
tions that the population growth rate has a negative influence on political stability. Thus,
as in Manarik (1981), any increase in population leads to a decrease in the stability of SSA
states. Similarly, we also show that natural resource abundance reduces political stability
and increases political instability in SSA, which brings us closer to the work of Dupasquier
and Osakwe (2006) on the natural resource curse. Similarly, remittances negatively influ-
ence the political stability of states in SSA, since, as in Avom et al. (2020), remittances
finance terrorism, which makes states fragile in terms of security and increases political
instability. Finally, for most of the results, foreign direct investment does not influence the
political stability of states. Moreover, we also find, as in Eggoh et al., (2021), that control-
ling corruption increases the political stability of states since it contributes to strengthen-
ing the legality of states. Finally, we show that wheat price volatility negatively influences
the political stability of states. Thus, as in Bellemare (2014) and Lagi et al. (2011), any
increase in the price of wheat will result in a decrease in the political stability of states. On
the other hand, our results show that the level of education reinforces the political stability
of states in SSA, better said as in Xu (2011), the level of education reduces political insta-
bility. Indeed, educated populations will be able to go through democratic and institutional
voices to claim rights, instead of going through protests that cause conflicts. Also, the rate
of economic growth does not influence the political stability of states in SSA.
Table 1  Direct effect of climate change on political stability in Africa
Independent Dependent Variable: Political stability index (PSI)
Variables
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

Estimator TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE
Variation of −0.0643*** −0.0383 −0.0381 −0.0502* −0.0919*** − 0.0906*** − 0.0916*** − 0.0938*** − 0.0936*** − 0.0943***
average (0.0248) (0.0275) (0.0275) (0.0273) (0.0269) (0.0264) (0.0264) (0.0265) (0.0264) (0.0264)
temperature
Variation of 0.0018** 0.0018** 0.0021*** 0.00193** 0.00275*** 0.00268*** 0.00251*** 0.00272*** 0.00275***
average (0.0008) (0.0008) (0.000824) (0.000795) (0.000793) (0.000795) (0.000821) (0.000861) (0.000861)
precipita-
tion
Gross domes- 0.00429 0.0103 0.00683 0.00922 0.00891 0.00927 0.00909 0.00953
tic product
per capita
(0.0072) (0.00735) (0.00710) (0.00699) (0.00699) (0.00700) (0.00704) (0.00704)
Population − 0.0260*** − 0.0293*** − 0.0342*** − 0.0354*** − 0.0355*** − 0.0337*** − 0.0336***
Climate change and political stability in African states: the…

grow (0.00596) (0.00576) (0.00572) (0.00580) (0.00581) (0.00610) (0.00610)

Natural − 0.0286*** − 0.0286*** − 0.0293*** − 0.0293*** − 0.0279*** − 0.0277***
resources (0.00345) (0.00339) (0.00344) (0.00344) (0.00388) (0.00388)
Remittance’s − 0.0254*** − 0.0258*** − 0.0262*** − 0.0246*** − 0.0246***
fund (0.00426) (0.00428) (0.00431) (0.00435) (0.00435)
Foreign direct 0.0552*** 0.00544 0.00597 0.00600
investment (0.00447) (0.00447) (0.00448) (0.00448)
Education 0.0135*** 0.00599*** 0.00518
(0.00158) (0.00172) (0.00172)
Control of 0.0763 0.0825
corruption (0.0671) (0.0672)
Volatility of 0.054***
food price (0.01)
(Wheat)
Constant − 0.438** − 0.62*** − 0.63*** − 0.527*** − 0.103 − 0.0182 − 0.0115 − 0.0376 − 0.0426 − 0.0404
(0.176) (0.197) (0.197) (0.196) (0.196) (0.193) (0.193) (0.196) (0.196) (0.196)
Table 1  (continued)
Independent Dependent Variable: Political stability index (PSI)
Variables
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

R square 0.109 0.113 0.114 0.132 0.192 0.222 0.224 0.224 0.233 0.234
Fisher 5.36*** 5.35*** 5.13*** 5.61 *** 5.41*** 8.74*** 9.77 *** 9.44*** 9.39*** 9.24*** (0.00)
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Number of 945 945 945 945 945 945 945 945 945 945
observa-
tions
Number of 45 45 45 45 45 45 45 45 45 45
countries

Standard errors are in parentheses

***
p < 0.01, ** p < 0.05, * p < 0.1 represent statistical significance
A. B. Henri Aurélien, O. N. Bruno Emmanuel
Climate change and political stability in African states: the…

5.2 Indirect effects

In Table 2, we perform an indirect analysis of the effect of climate change on political sta-
bility through the channel of food and agricultural price volatility. Indeed, as suggested by
Berazneva and Lee (2013) and Koubi et al. (2012), rising food and agricultural prices are
observed but rarely used as intermediate variables. However, if for Macauley and Rama-
djita (2015) the production of wheat in SSA is lower than that of maize, and rice, studies
such as those of Nelson et al. (2009) indicate unlike these crops, wheat is the crop most
vulnerable to the effects of climate change. They estimate that by 2050 in sub-Saharan
Africa, declines in rice, wheat, and maize yields will be 15%, 34%, and 10% respectively
due to climate change. Furthermore, we are particularly interested in the price of wheat,
whose rise from US$150 per ton to US$340 between September 2005 and September 2007
led to the riots and conflicts of 2008 (Dufumier & Hugon, 2008). Furthermore, the World
Bank (2022) indicates that due to the war in Ukraine, world wheat prices have increased
from 48 to 79% between 2021–2022, thus aggravating food insecurity and political ten-
sions in West Africa and Sahel regions for example.
Primarily, from the interactions between climate change and wheat price volatility (Eq. 11),
we find like Berazneva and Lee (2013) and Koubi et al. (2012), that there is an indirect effect
of climate change on political stability in SSA. Specifically, from the interaction between
wheat price volatility and rainfall abundance, we find that an increasingly volatile wheat price
will reduce the significant and positive effect of rainfall abundance on political stability, and
may then create instability. On the other hand, concerning the interaction between wheat price
volatility and temperature levels, we find that higher wheat prices will aggravate the significant
negative effect of higher temperatures on political stability. Moreover, introducing wheat price
volatility as a transmission channel does not change our main results. Rising temperatures are
still likely to decrease political stability while rising rainfall makes states even more stable.
Similarly, the effect of control variables is identical for almost all variables.

6 Robustness and sensitivity test

In this article, we propose three robustness tests and one sensitivity test.

6.1 Robustness tests

The first two robustness tests highlight other methods and estimation techniques to control
for endogeneity problems, while the third identifies another measure of political stability.

6.1.1 Taking endogeneity into account by lagged explanatory variables

Considering the use of several interdependent climatic and non-climatic explanatory

variables, we take into account the problem of endogeneity of the control variables to
guarantee the robustness of our results (Yeyuomo and Asongu, 2023). To achieve this, we
perform panel regressions using the TWFE, with explanatory variables lagged by one or
two periods to avoid problems of simultaneity, multicollinearity, and overparameterization
(Bellemare et al., 2017; Asongu & Nwachukwu, 2018). In addition, Gnimassoun and
Santos (2021) show that the use of this method makes it possible to mitigate the problem
posed by the two-step least squares (2SLS) method, in particular, that linked to the
 A. B. Henri Aurélien, O. N. Bruno Emmanuel

Table 2  Indirect effect of climate change on political stability: wheat price volatility as a transmission
channel
Dependent Variable: Political stability index (PSI)
Independent Variables (11)

Estimator TWFE
Variation of average temperature − 0.0732***
(0.009)
Variation of average precipitation 0.00603***
(0.00129)
Gross domestic product per capita 0.0991***
(0.00706)
Population grow − 0.0341***
(0.00580)
Natural resources − 0.0294***
(0.00342)
Remittance’s fund − 0.0239***
(0.00431)
Foreign direct investment 0.00400
(0.00446)
Education 0.00736***
(0.00170)
Corruption control 0.0745
(0.0667)
food of price volatility (wheat) − 0.00562***
(0.00127)
food of price volatility (wheat) * Variation of average temperature − 0.00151***
(0.000344)
food of price volatility (wheat) * Variation of average precipitation − 0.0087**
(0.0029)
Constant − 0.546**
(0.225)
R square 0.2
Fisher 9.44***
(0.00)
Number of observations 937
Number of observations 45

Standard errors are in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1 represent statistical significance

subjectivity of the choice of the relevant instruments. The results contained in Table 3 (for
the variables lagged by one period) and in Appendix 7 (for those lagged by two periods)
are consistent with the previous results. Our results, as presented in Table 3 (Eq. 12–21),
remain globally unchanged, thus confirming their robustness.

6.1.2 Taking endogeneity into account using the GMM‑S

Omission of relevant variables and measurement error are the main endogeneity
biases that may exist between climate change and political stability, given the purely
Climate change and political stability in African states: the…

exogenous nature of climate variables. Dynamic panel models are confronted with the
correlation between unobservable country-specific effects and the lagged dependent
variable, which leads to inconsistent estimators under ordinary least squares (OLS).
Therefore, Arellano and Bond (1991) developed a consistent estimator using the lagged
values of the first difference of the endogenous variable as an instrument, called GMM
in difference (GMM-D). However, due to the persistence of the dependent variable,
this estimator makes the lagged values very poor instruments (Blundell & Bond,
1998). Using additional moment conditions, Blundell and Bond (1998) proposed a
more robust alternative estimator called the GMM-S, which is based on a system of
two equations, one in level and the other in first difference.
We then use the generalized system method of moments (GMM-S) developed by
Blundell and Bond (1998). Leitão and Shahbaz (2016) indicate that, like fixed effects,
the GMM-S estimator makes it possible to control potential endogeneity problems.
Indeed, extending the approach developed by Arellano and Bover (1995), Rood-
man (2009) shows that the GMM-S estimator is biased when the estimation strategy
imposes too many instruments, and overcomes this problem by limiting the number
of instruments and maximising the sample size using the direct orthogonal deviation
technique. For Baum et al. (2003), the GMM-S corrects for endogeneity, heteroske-
dasticity and autocorrelation of the errors. In our study, GMM-S makes it possible to
correct for omitted variable bias through the use of instrumental variables (Hill et al.,
2021).
Specifically, this method allows us to: (i) to take into account unobservable het-
erogeneous factors specific to each SSA country, such as cultural, historical, political,
economic, geographical, or institutional factors; (ii) to correct the double simultaneity
that may exist between the political stability, climate change, and all the explanatory
variables; (iii) to correct any endogeneity problem arising from a potential correlation
between the error term and the lagged endogenous variable. Thus, to obtain an unbi-
ased, efficient, and consistent estimator, we use, on the one hand, instrumental vari-
ables delayed by at least two periods, such as GDP per capita, the population growth
rate, natural resources, and the volatility of food prices considered as endogenous.
Moreover, our panel validates the GMM-S hypothesis relating to the superiority of
individuals (N = 45) over the period (T = 21) (Roodman, 2009).
The use of GMM-S as a robustness measure is validated by the Sargan test of instru-
ment validity (Table 4, Eq. 22–31). Furthermore, we again arrive at the same results as
before. In other words, rising temperatures decrease political stability, while abundant
rainfall contributes to the stabilization of states in SSA. Furthermore, the results on the
influence of control variables on political stability are not changed.
Also, the robustness of our results by the GMM-S is validated by the hypothesis of
autocorrelation of order 1 and the absence of autocorrelation of order 2, as well as by
the significance of the Sargan test.

6.2 Rule of law as another measure of the political stability of states

To analyze the effect of climate change on political stability, we perform robustness

test at this level using the World Governance Indicator’s rule of law (WGI, 2020), as
an alternative measure of states’ political stability. Indeed, as Radu (2015) states, it has
Table 3  Taking endogeneity into account by lagged explanatory variables
Independent Dependent Variable: Political stability index (PSI)
Variables
(12) (13) (14) (15) (16) (17) (18) (19) (20) (21)

Estimator TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE
Variation of − 0.0600** − 0.0346 − 0.0346 − 0.0458* − 0.0853*** − 0.0844*** − 0.0854*** − 0.0867*** − 0.0851*** − 0.0859***
average tem-
peratureit-1
(0.0248) (0.0275) (0.0275) (0.0274) (0.0270) (0.0265) (0.0265) (0.0267) (0.0268) (0.0268)
Variation of 0.0017** 0.001** 0.002** 0.00190** 0.00275*** 0.00268*** 0.00258*** 0.00294*** 0.00299***
average pre-
cipitationit-1
(0.0008) (0.0008) (0.0008) (0.000800) (0.000798) (0.000800) (0.000826) (0.000875) (0.000875)
Gross domestic 0.00054 0.00610 0.00280 0.00521 0.00490 0.00510 0.00335 0.00379
product per
capitait-1
(0.0072) (0.0073) (0.00715) (0.00703) (0.00703) (0.00705) (0.00715) (0.00715)
Population − 0.024*** − 0.0273*** − 0.0322*** − 0.0334*** − 0.0335*** − 0.0314*** − 0.0312***
growit-1
(0.00597) (0.00579) (0.00575) (0.00584) (0.00584) (0.00620) (0.00619)
Natural − 0.0271*** − 0.0272*** − 0.0279*** − 0.0279*** − 0.0256*** − 0.0255***
resourcesit-1
(0.00348) (0.00341) (0.00346) (0.00346) (0.00394) (0.00394)
Remittance’s − 0.0256*** − 0.0260*** − 0.0263*** − 0.0258*** − 0.0258***
fundit-1
(0.00431) (0.00432) (0.00436) (0.00443) (0.00443)
Foreign direct 0.00540 0.00536 0.00517 0.00521
investmentit-1
(0.00450) (0.00450) (0.00455) (0.00455)
Educationit-1 0.000752 − 0.000279 − 0.000371
(0.00158) (0.00174) (0.00174)
Control of cor- 0.0745 0.0811
ruptionit-1
A. B. Henri Aurélien, O. N. Bruno Emmanuel

(0.0682) (0.0684)
Table 3  (continued)
Independent Dependent Variable: Political stability index (PSI)
Variables
(12) (13) (14) (15) (16) (17) (18) (19) (20) (21)

Volatility of 0.042*
food price
(Wheat)it-1
(0.021)
Constant − 0.417** − 0.61*** − 0.61*** − 0.512** − 0.168 − 0.0980 − 0.0999 − 0.130 − 0.0893 − 0.0841
(0.178) (0.200) (0.202) (0.202) (0.201) (0.197) (0.197) (0.207) (0.210) (0.210)
R square 0.108 0.112 0.112 0.128 0.182 0.213 0.214 0.214 0.216 0.218
Fisher 5.31***(0.00) 5.30***(0.00) 5.06***(0.00) 5.61***(0.00) 8.12***(0.00) 9.52***(0.00) 9.23 ***(0.00) 8.90***(0.00) 8.60***(0.00) 8.39***(0.00)
Number of 943 943 943 943 943 943 943 943 936 936
observations
Number of 45 45 45 45 45 45 45 45 45 45
countries
Climate change and political stability in African states: the…

Standard errors are in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1 represent statistical significance
Table 4  GMM-S as another estimator
Dependent Variable: Political stability index (PSI)
(22) (23) (24) (25) (26) (27) (28) (29) (30) (31)

Political stability index (-1) 0.957*** 0.748*** 0.700*** 0.621* 0.938*** 1.056*** 0.346* 1.267*** 0.553*** 0.462**
(0.0908) (0.206) (0.118) (0.348) (0.168) (0.131) (0.179) (0.203) (0.0751) (0.196)
Variation of average temperature − 0.123* − 0.294 − 0.145* − 0.275 − 0.680*** − 0.413*** − 0.248** − 0.430*** − 0.394*** − 0.389*
(0.0664) (0.241) (0.0835) (0.510) (0.148) (0.113) (0.121) (0.148) (0.0750) (0.209)
Variation of average precipitation 0.0131* 0.000382 0.0196** 0.0131*** 0.00385* 0.0169** 0.00177 0.00468* 0.00129
(0.00763) (0.00169) (0.00926) (0.00375) (0.00199) (0.00811) (0.00339) (0.00265) (0.00565)
Gross domestic product per capita 0.0233*** 0.0343 0.0538*** 0.0418*** 0.0557* 0.0658*** 0.0104 0.0193
(0.00864) (0.0278) (0.0155) (0.0120) (0.0330) (0.0250) (0.0163) (0.0334)
Population grow − 0.0364* − 0.0251*** − 0.00483 − 0.0391*** − 0.0228*** − 0.0423*** − 0.0613**
(0.0196) (0.00679) (0.00674) (0.0145) (0.00789) (0.0125) (0.0238)
Natural resources − 0.02*** − 0.013*** − 0.0129* − 0.0196** − 0.0411*** − 0.0419***
(0.00713) (0.00471) (0.00662) (0.00917) (0.00859) (0.0133)
Remittance’s fund − 0.0704** 0.0245 − 0.0246*** − 0.0204 − 0.0461
(0.0279) (0.0331) (0.00892) (0.0302) (0.0395)
Foreign direct investment − 0.0525 − 0.0786 − 0.000767 − 0.000591
(0.0432) (0.0521) (0.0145) (0.0314)
Education − 0.0206*** − 0.0155*** − 0.0158*
(0.00775) (0.00577) (0.00927)
Control of corruption 0.512*** − 0.725*
(0.190) (0.375)
Volatility of food price (Wheat) 0.01***
(0.012)
Constant 0.252 1.530 − 0.485 1.964 2.962*** 1.124** 1.800** 2.331*** 2.081*** 1.687**
(0.191) (1.065) (0.322) (2.002) (0.647) (0.445) (0.904) (0.627) (0.510) (0.765)
AR(1) 0.07 0.05 0.03 0.01 0.01 0.02 0.05 0.02 0.01 0.02
A. B. Henri Aurélien, O. N. Bruno Emmanuel
Table 4  (continued)
Dependent Variable: Political stability index (PSI)
(22) (23) (24) (25) (26) (27) (28) (29) (30) (31)

AR(2) 0.14 0.25 0.25 0.12 0.55 0.42 0.17 0.21 0.55 0.18
Sargan p-value 0.35 0.42 0.54 0.11 0.67 0.33 0.23 0.28 0.41 0.13
Instruments 15 17 19 12 16 23 25 23 27 29
Number of observations 269 269 269 269 269 269 269 269 269 269
Number of countries 45 45 45 45 45 45 45 45 45 45

Standard errors are in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1 represent statistical significance
Climate change and political stability in African states: the…
 A. B. Henri Aurélien, O. N. Bruno Emmanuel

been shown that countries with a strong rule of law are characterized by low levels of
political unrest.
For all specifications (Table 5, Eq. 31–41), we find that when the rule of law is used
as an endogenous variable of political stability, our main results are not changed. In
other words, rising temperatures worsen political stability, while abundant rainfall con-
tributes to the political stability of states. On the other hand, the results of the influ-
ence of the control variables on political stability are not changed.

6.3 Sensitivity test: introduction of historical variable into the analysis

The political instability of African countries is strongly linked to the fact that they were colo-
nized (Feeny et al., 2016). Furthermore, the European University Institute (2009) notes that
the artificial nature of colonial institutions, the development of political dependence on the
colonial power, and the system of indirect administration, have contributed to the instability
of many African states. Similarly, Fenske (2010) shows that the failure of the colonial powers
to develop autonomous institutions has had an overwhelming influence on the emergence and
prevalence of poverty in post-colonial sub-Saharan Africa. Therefore, to test the sensitivity of
our results, we introduce a historical variable called colonization. The colonization variable is
measured by legal origin (French) from the database of La porta et al. (1999).
Our results in Table 6 above show that, colonization increases political instability in
SSA as well as in the colonized (Feeny et al., 2016) for most specifications (Eq. 42–47).
Indeed, as shown by N’Dimina-Mougala (2007), independence, secessionist, border and
identity conflicts have a common origin in colonization. Moreover, the introduction of the
colonization variable does not change our main results.

7 Discussion of main results and implications

This study is of capital importance insofar as the results obtained support previous work
on the direct effect of climate change on the deterioration of the political stability of States.
Indeed, these results, like most empirical work (Raleigh & Kniveton, 2012; Raleigh et al.,
2015) on the issue, indicate that any increase in temperature levels reduces the political
stability of states. Furthermore, and in relation to other works (by Lecoutere et al., 2010;
Raleigh & Urdal, 2007) the abnormal drop in precipitation leads to a reduction in the stability
of States. The importance and credibility of this result is based on the neo-Malthusian theory
of resource scarcity (Koubi, 2019). Indeed, climate change, by reducing natural resources
such as fresh water, arable land, forests and fisheries, leads to a reduction in policy. The WMO
report (2020) indicates that water scarcity due to climate change is the cause of tensions and
conflicts in the region. For example, the scarcity of water resources due to climate change
has generated clashes between Arab Choua herders and Mousgoum farmers in the Far North
region of Cameroon and in the desert areas of the Sahel. The consequences of this scarcity of
resources are felt on agricultural production which, according to experts, has seen a decline
of 40% over the last 10 years. Indeed, the agricultural sector is at the heart of the economies
of States, since it contributes to 23% of GDP and constitutes a real source of employment,
and represents on average 55% of the active population of these countries according to the
International Organization work.International Labour Office (2017). Thus, any disruption of
agricultural production due to climate change is a source of conflict and protests. However,
Table 5  Direct effects: another measure of political stability
Dependent Variable: Rule of law index (RLI)
(32) (33) (34) (35) (36) (37) (38) (39) (40) (41)

Estimator TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE
Variation of 0.0351** 0.0359** 0.0360** 0.0306* − 0.00393 − 0.00393 − 0.00436 − 0.00828 − 0.000754 − 0.00107
average
temperature
(0.0143) (0.0159) (0.0159) (0.0159) (0.0151) (0.0151) (0.0147) (0.0147) (0.0143) (0.0143)
Variation of 0.00108** 0.00108** 0.00122** 0.00100** 0.00100** 0.00146*** 0.00113** 0.00241*** 0.00242***
average
precipita-
tion
(0.000481) (0.000482) (0.000481) (0.000449) (0.000449) (0.000445) (0.000456) (0.000465) (0.000465)
Gross domes- 0.00207 0.00468 0.00206 0.00206 0.00336 0.00405 − 0.000744 − 0.000535
tic product
per capita
Climate change and political stability in African states: the…

(0.00420) (0.00425) (0.00397) (0.00397) (0.00388) (0.00387) (0.00379) (0.00379)

Population − 0.0113*** − 0.0140*** − 0.0140*** − 0.0181*** − 0.0184*** − 0.00906*** − 0.00900***
grow
(0.00345) (0.00322) (0.00322) (0.00322) (0.00321) (0.00328) (0.00328)
Natural − 0.0228*** − 0.0228*** − 0.0234*** − 0.0231*** − 0.0147*** − 0.0146***
resources
(0.00194) (0.00194) (0.00191) (0.00191) (0.00209) (0.00209)
Remittance’s − 0.0166*** − 0.0175*** − 0.0137*** − 0.0137***
fund
(0.00237) (0.00238) (0.00234) (0.00234)
Foreign direct 0.00442* 0.00424* 0.00218 0.00220
investment
(0.00248) (0.00247) (0.00241) (0.00241)
Education 0.00270*** − 0.000546 − 0.000583
Table 5  (continued)
Dependent Variable: Rule of law index (RLI)
(32) (33) (34) (35) (36) (37) (38) (39) (40) (41)

(0.000875) (0.000927) (0.000927)

Control of 0.306*** 0.309***
corruption
(0.0361) (0.0362)
Volatility of − 0.020***
food price
(Wheat)
(0.008)
Constant − 0.807*** − 0.922*** − 0.924*** − 0.877*** − 0.533*** − 0.533*** − 0.470*** − 0.526*** − 0.487*** − 0.486***
(0.102) (0.114) (0.114) (0.114) (0.110) (0.110) (0.108) (0.109) (0.106) (0.106)
R square 0.618 0.620 0.620 0.625 0.674 0.674 0.691 0.695 0.694 0.695
Fisher 67.74*** 64.75*** 63.17*** 75.70 *** 75.30*** 75.30*** 75.28 *** 73.61*** 70.27*** 68.01***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Number of 935 935 935 935 935 935 935 935 928 928
observa-
tions
Number of 45 45 45 45 45 45 45 45 45 45
countries

Standard errors are in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1 represent statistical significance
A. B. Henri Aurélien, O. N. Bruno Emmanuel
Table 6  Climate change and political stability: introducing a historical variable
Dependent Variable: Political stability index (PSI)
(42) (43) (44) (45) (46) (47) (48) (49) (50)

Estimator TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE
Variation of − 0.0547** − 0.00627 − 0.00614 − 0.0170 − 0.0537** − 0.0512** − 0.0522** − 0.0572** − 0.0569**
average tem-
perature
(0.0217) (0.0242) (0.0242) (0.0240) (0.0236) (0.0232) (0.0232) (0.0232) (0.0229)
Variation of 0.0032*** 0.00321*** 0.00354*** 0.00331*** 0.00409*** 0.00402*** 0.00358*** 0.00378***
average pre-
cipitation
(0.000732) (0.000732) (0.000727) (0.000702) (0.000700) (0.000702) (0.000718) (0.000748)
Gross domestic 0.00192 0.00763 0.00483 0.00710 0.00683 0.00782 0.00770
product per
capita
Climate change and political stability in African states: the…

(0.00630) (0.00634) (0.00613) (0.00602) (0.00603) (0.00602) (0.00602)

Population − 0.0246*** − 0.0275*** –0.0320*** − 0.0330*** − 0.0334*** − 0.0316***
grow
(0.00514) (0.00498) (0.00493) (0.00501) (0.00499) (0.00522)
Natural − 0.0244*** –0.0245*** − 0.0251*** − 0.0248*** − 0.0235***
resources
(0.00299) (0.00293) (0.00297) (0.00296) (0.00332)
Remittance’s − 0.0227*** − 0.0230*** − 0.0243*** − 0.0227***
fund
(0.00369) (0.00370) (0.00371) (0.00373)
Foreign direct 0.00470 0.00445 0.00498
investment
(0.00385) (0.00384) (0.00383)
Education 0.00373*** 0.00303**
(0.00136) (0.00147)
Table 6  (continued)
Dependent Variable: Political stability index (PSI)
(42) (43) (44) (45) (46) (47) (48) (49) (50)

Control of cor- 0.0724

ruption
(0.0574)
Volatility of − 0.0348*
food price
(Wheat)
(0.0811)
Colonisation − 0.339** − 0.238* − 0.236* − 0.230* − 0.220* − 0.216* − 0.134 − 0.117 − 0.114
(0.139) (0.139) (0.139) (0.138) (0.133) (0.131) (0.131) (0.131) (0.129)
Constant − 0.137 − 0.577** − 0.580** − 0.508** − 0.153 − 0.145 − 0.135 − 0.229 − 0.238
(0.206) (0.228) (0.228) (0.226) (0.222) (0.218) (0.218) (0.220) (0.218)
R square 0.141 0.159 0.159 0.180 0.236 0.267 0.268 0.274 0.285
Fisher 6.83*** (0.00) 7.50*** (0.00) 7.98 *** (0.00) 10.79*** (0.00) 12.22 *** 11.84 *** 11.77*** 11.90*** 11.64*** (0.00)
(0.00) (0.00) (0.00) (0.00)
Number of 935 935 935 935 935 935 935 935 928
observations
Number of 45 45 45 45 45 45 45 45 45
countries

Standard errors are in parentheses, ***p < 0.01, **p < 0.05, *p < 0.1 represent statistical significance
A. B. Henri Aurélien, O. N. Bruno Emmanuel
Climate change and political stability in African states: the…

climate disruptions are not simply limited to agricultural production, but also extend to
the structure of agricultural prices. Indeed, considered one of the agricultural products
most vulnerable to climate change (Nelson et al., 2009), the volatility of the price of wheat
constitutes one of the mechanisms by which climate change affects political stability. In our
study, we tested this hypothesis based on the interactions between climate change and wheat
price volatility, following the work of Berazneva and Lee (2013) and Koubi et al. (2012). We
found that an increasingly volatile wheat price reduces the significant and positive effect of
rainfall abundance on political stability. On the other hand, regarding the interaction between
wheat price volatility and temperature levels, we showed that rising wheat prices aggravate
the significant negative effect of rising temperatures on political stability. The resurgence
of climate shocks in recent decades, and the hunger riots of 2008 linked to the rise in the
price of wheat, are palpable examples of the role that the interconnection between climate
change and the volatility of agricultural prices can play in fueling conflicts. It emerges that
the establishment of climate change adaptation strategies at the national level (through
government policies), and at the international level (via bilateral agreements) are the ways
by which climate conflicts can be anticipated. These may include innovations in agricultural
production techniques more adapted to extreme climatic conditions, such as the creation of
water storage points, which would make it possible to irrigate fields, and the establishment
of a food storage system to cope with soaring prices. Furthermore, these discussions and
proposals are validated by several robustness and sensitivity tests which do not change the
meaning of the results suggested in this study.

7.1 Conclusion and recommendations

This paper aims to analyze the effect of climate change on the political stability of coun-
tries in sub-Saharan Africa. First, the direct effect of climate change on political stability
was examined. Secondly, we analyzed the indirect effect of climate change on political sta-
bility through wheat price volatility. To achieve these objectives, we started with a sample
of 45 African countries over a period from 2000 to 2020. The model used is the two-factor
fixed effects model, with the Two-way fixed effect as an estimator, appropriate for taking
into account the problem of endogeneity bias.
The estimation results led us to two main conclusions: first, climate change reduces the
political stability of SSA states. Specifically, we have shown that climate change reduces
political stability through moments of crisis or conflict in relatively warm periods. In addi-
tion, abundant rainfall strengthens political stability in SSA because these are economies
where GDP is agricultural and agriculture contributes to employment. Conversely, any
decrease in rainfall leads to a decrease in political stability. These results are explained by
the mechanism of resource scarcity, which affects the living conditions of the population. In
fact, increasingly dry periods without rainfall reduce the means of subsistence of populations
whose crops depend on the climate. This vulnerability will therefore lead them to turn to
the government for adaptation or mitigation strategies related to climate change. As a result,
the low resilience of leaders in SSA to provide adequate solutions to the population reduces
their legitimacy, which contributes to the decline of political stability. The political instability
specific to SSA countries, leading to the overthrow of a government (legitimately or through
coup d’Etat), is exacerbated by the impact of climate change on populations. The hunger riots
in 2008 due to the high cost of living and the increase in the price of basic food and agri-
cultural commodities are evidence of the link between climate, agricultural production and
political instability. The second result allowed us to strengthen our argument by showing that
 A. B. Henri Aurélien, O. N. Bruno Emmanuel

climate change affects the political stability of states through the volatility of wheat prices.
Thus, using marginal effects, we showed that an increasingly volatile wheat price would
reduce the significant and positive effect of abundant rainfall on political stability and could
then lead to instability. On the other hand, we found that rising wheat prices would exacer-
bate the negative and significant effect of rising temperatures on political stability.
The analysis of the control variables showed that natural resources, population
growth rate, remittances, and wheat price volatility hurt political stability on the one
hand. On the other hand, the level of education and the control of corruption rein-
force the political stability of states. However, with the GMM-S estimator and lagged
explatonary variables method as robustness, we took into account endogeneity prob-
lems, and validated the robustness of our main results. To reduce the risks of instability
in Africa, we need to set up effective methods for adapting to or mitigating the effects of
climate change, by preserving the environment and promoting green energies that emit
less CO2. Specifically, we need to put in place measures to mitigate or adapt the harm-
ful effects of climate change. Thus, in times of drought and the face of reduced rain-
fall, anticipating fragility means, for example, creating and/or reinforcing water man-
agement and storage infrastructures that meet the needs of populations, farmers, and
stockbreeders, so as not to limit subsistence. Also, we need to integrate local develop-
ment policies and programs that train people to resist and adapt quickly and cheaply to
disasters caused by climate change. Finally, we need to implement measures to stabilise
food prices before, during, and after periods of political instability. In concrete terms,
stabilising food price volatility means, for example, increasing the level of national and
regional stocks of various foodstuffs, limiting or delaying the influence of a climatic
shock such as drought, limiting the production of agrofuels, by ensuring national access
to better information on the level of food prices on international markets.
However, our study has a certain number of limitations that should be noted. The scope of
our study in Sub-Saharan Africa is our main limitation. Indeed, the issues of climate change
and political stability are global and are not limited only to SSA. Just like this area, several
regions in Europe, Asia and America are suffering from the ravages of climate change and
political instability. The second limitation is based on the horizon of the study. Indeed, the
study period is limited to 21 years due to the lack of data, particularly those related to politi-
cal stability for several countries. Thus, future perspectives would consist on the one hand of
extending our study to a larger number of countries (for all continents) and over a longer time
horizon, to better analyze the long-term phenomena of climate change and political stability.

Appendix 1

Country of our sample

Angola, Benin, Burkina Faso, Botswana, Burundi, Cameroon, Cape Verde, Central Afri-
can Republic, Chad, Comoros, Congo-Brazzaville, Djibouti, Equatorial Guinea, Eritrea,
Ethiopia, Gabon, Gambia, Ghana, Guinea-Bissau, Guinea, Cote d’Ivoire, Kenya, Liberia,
Lesotho, Mauritania, Madagascar, Mauritius, Malawi, Mali, Mozambique, Namibia, Nige-
ria, Niger, Rwanda, South Africa, Senegal, Sierra Leone, Somalia, Sudan, Tanzania, Togo,
Uganda, Dem, Rep, of Congo, Zambia and Zimbabwe.
Climate change and political stability in African states: the…

Appendix 2

Variables, descriptions and data sources. Source Authors

Variables Source data

Political stability index (PSI) World Governance Indicator (WGI, 2020)

Rule of law World Governance Indicator (WGI, 2020)
Variation of average temperature (VAT) World Bank Group Climate Change
Knowledge Portal (CCKP, 2021)
Variation of average precipitation (VAP)
Food price volatility (VOFP) of wheat FAO (Faostat, 2020)
Control of corruption World Governance Indicator (WGI, 2020)
Economic growth rate (GDP) World development Indicator (WDI, 2020)
Education (Educ) World Development Indicator (WDI, 2020)
Demography (POPGR)
Natural ressources ( Natres)
Remittances fund (Remf)
FDI
Colonialisation (col) Author according to literature

Appendix 3

Means and standard deviation of variables

Variable Obs Mean Std. dev Min Max

Political stability 935 −.5297575 1.02861 −3.314937 2.4

VAP 945 74.12822 48.39955 9.915553 361.4276
VAT 945 2.292662 1.455645 .406761 7.972619
GDPCAP 945 1.564324 5.191211 −36.55692 56.78894
POPGRO 945 3.478749 6.202973 −.5670208 62.03992
NATRES 945 10.93712 10.35344 .2652482 62.69702
REMF 945 4.808855 8.134079 .0001832 57.50477
CORRUPCO 938 −.6698951 .6639798 −1.868714 7
EDUC 945 44.66217 23.90447 −21.70256 114.5126
vofpwheat 945 202.4086 2384.908 .91 69435.37
FDI 945 4.513236 7.934918 −11.19897 103.3374
col 117 .9555556 .2061685 0 1
Appendix 4

Correlation matrix between variables

Political VAP VAT GDPCAP POPGRO NATRES REMF CORRUPCO EDUC vofpblé FDI col
stability

Political
stability 1.0000
VAP 0.1546 1.0000
VAT − 0.0667 − 0.4392 1.0000
GDPCAP 0.0475 0.0096 − 0.0170 1.0000
POPGRO − 0.1079 0.1457 − 0.1557 0.1860 1.0000
NATRES − 0.2227 0.0432 − 0.1807 − 0.0310 − 0.0498 1.0000
REMF − 0.1238 0.1640 − 0.0554 0.0265 − 0.1093 0.0083 1.0000
CORRUPCO 0.1986 − 0.2627 0.1942 0.0786 − 0.2297 − 0.4109 − 0.1021 1.0000
EDUC 0.0479 0.2369 − 0.0305 − 0.0915 0.0095 − 0.0560 0.1491 0.2305 1.0000
vofpwheat 0.0338 − 0.0164 0.0337 − 0.0490 − 0.0160 0.0132 0.0121 − 0.0496 0.0047 1.0000
FDI − 0.0297 0.1092 − 0.0665 0.0660 0.1717 0.1593 0.0717 − 0.0493 0.0540 − 0.0174 1.0000
col − 0.0675 − 0.1036 − 0.1053 − 0.0319 0.0338 0.0283 0.0534 − 0.0300 − 0.0625 0.0128 0.0269 1.0000
A. B. Henri Aurélien, O. N. Bruno Emmanuel
Appendix 5

VIF (variance of inflation factor) test of multicollinearity

Variable VIF 1/VIF

VAT 8.02 0.124964

VAP 3.66 0.273221
GDPCAP 1.46 0.685331
POPGRO 5.21 0.191938
NATRES 5.09 0.196423
REMF 8.66 0.115464
CORRUPCO 6.16 0.162295
EDUC 7.71 0.129706
FDI 1.56 0.640606
vofpwheat 1.12 0.893662
Climate change and political stability in African states: the…

Appendix 6

Breusch-Godfrey LM test for autocorrelation. Source Authors

lags(p) chi2 df Prob > chi2

1 2.314 1 0.1282
Appendix 7

Breush-Pagan heteroskedasticity test

lags(p) chi2 df Prob > chi2

1 0.037 1 0.8469

Appendix 8

Taking endogeneity into account by lagged explanatory variables

Independent Dependent Variable: Political stability index (GFI)

Variables
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

Estimator TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE TWFE
Variation of − 0.0564** − 0.0368 − 0.0370 − 0.0505* − 0.0884*** − 0.0875*** − 0.0890*** − 0.0896*** − 0.0868*** − 0.0877***
average tem-
peratureit-2
(0.0247) (0.0274) (0.0274) (0.0272) (0.0269) (0.0264) (0.0264) (0.0265) (0.0266) (0.0265)
Variation of 0.00137* 0.00137* 0.00174** 0.00157** 0.00242*** 0.00231*** 0.00227*** 0.00290*** 0.00295***
average pre-
cipitationit-2
(0.0008) (0.0008) (0.000820) (0.000796) (0.000794) (0.000795) (0.000821) (0.000867) (0.000867)
Gross domestic − 0.00346 0.00326 0.000156 0.00271 0.00223 0.00232 − 0.000511 − 2.72e− 05
product per
capitait-2
(0.0072) (0.00733) (0.00713) (0.00701) (0.00700) (0.00702) (0.00710) (0.00710)
A. B. Henri Aurélien, O. N. Bruno Emmanuel
Independent Dependent Variable: Political stability index (GFI)
Variables
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

Population − 0.0292*** − 0.0321*** − 0.0371*** − 0.0390*** − 0.0391*** − 0.0350*** − 0.0348***

growit-2
(0.00593) (0.00577) (0.00572) (0.00580) (0.00581) (0.00615) (0.00614)
Natural − 0.0260*** − 0.0261*** − 0.0272*** − 0.0272*** − 0.0233*** − 0.0231***
resourcesit-2
(0.00346) (0.00340) (0.00344) (0.00344) (0.00392) (0.00391)
Remittance’s − 0.0258*** − 0.0265*** − 0.0266*** − 0.0251*** − 0.0251***
fundit-2
(0.00429) (0.00429) (0.00433) (0.00440) (0.00440)
Foreign direct 0.00859* 0.00857* 0.00795* 0.00798*
investmentit-2
(0.00447) (0.00447) (0.00451) (0.00451)
Educationit-2 0.000326 − 0.00148 − 0.00158
Climate change and political stability in African states: the…

(0.00157) (0.00173) (0.00173)

Control of cor- 0.136** 0.144**
ruptionit-2
(0.0677) (0.0678)
Volatility of 0.021**
food price
(Wheat)it-2
(0.009)
Constant − 0.430** − 0.580*** − 0.576*** − 0.490** − 0.192 − 0.126 − 0.125 − 0.139 − 0.0439 − 0.0405
Independent Dependent Variable: Political stability index (GFI)
Variables
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

(0.177) (0.198) (0.199) (0.197) (0.195) (0.192) (0.192) (0.202) (0.206) (0.206)
R square 0.113 0.116 0.116 0.139 0.189 0.220 0.223 0.223 0.228 0.230
Fisher 5.61*** 5.49*** 5.26*** 6.17 *** 8.54*** 9.93*** 9.72 *** 9.37*** 9.23*** 9.02***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Number of 942 942 942 942 942 942 942 942 935 935
observations
Number of 45 45 45 45 45 45 45 45 45 45
countries

Standard errors are in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1 represent statistical significance
A. B. Henri Aurélien, O. N. Bruno Emmanuel
Climate change and political stability in African states: the…

Author contributions The authors’ contributions are equal in all areas.

Data availability All data and material used are in the public domain.

Declarations
Conflict of interest The author does not declare anything.

Ethical approval This article uses publicly available data and therefore ethics approval is not required.

References
Alesina, A. (1989). Politics and business cycles in industrial democracies. Economic Policy, 4(8), 55–98.
Alesina, A., & Perotti, R. (1996). Revenue distribution, political instability and investment. European
Economic Review, 40, 1203–1228.
Anderson, C., & Bushman, B. (2002). Human aggression. Annual Review of Psychology, 53, 27–15.
Arellano, M., & Bond, S. (1991). Some tests of specification for panel data: Monte Carlo evidence and an
application to employment equations. The Review of Economic Studies, 58(2), 277.
Arellano, M., & Bover, O. (1995). Another look at the instrumental variable estimation of error-components
models. Journal of Econometrics, 68(1), 29–51.
Asongu, S. A., & Nwachukwu, J. C. (2019). The role of openness in the effect of ICT on governance.
Information Technology for Development, 25(3), 503–531.
Ateba, B. H., Ongo, N. B. O., Mougnol, A. E., & Mamadou, A. (2023). Does climate change influence
conflicts ? Evidence for the Cameroonian regions. GeoJournal, 88(4), 0123456789. https://​doi.​org/​
10.​1007/​s10708-​023-​10824-7
Ateba Boyomo, H. A., Ongo Nkoa, B. E., & Awah Manga, L. A. (2024). Climate change and livestock
production in Sub‐Saharan Africa: Effects and transmission channels. Food and Energy Security,
13(1), e521.
Avom, D., Ongo Nkoa, B. E., & Song, J. S. (2020). Les transferts de fonds des migrants (TFM) expliquent-
ils les fragilités des États africains? Revue Française D’économie, 35(4), 197–243.
Bachler, G., & Spillman, K. (1996). Environmental degradation as a cause of war. ENCOP-Final Report, 1.
Ballentine, K., & Nitzschke, H. (2003). Beyond greed and grievance: Policy lessons from studies in the
political economy of armed conflict. Stockholm: International Peace Academy.
Bank, W. (2022). The war in Ukraine - amplifying an already prevailing food crisis in West Africa and the
Sahel region.
Bannon, I., & Collier, P. (2003). Natural Ressources and Violent Conflict: Options and Actions. The World
Bank.
Baum, C. F., Schaffer, M. E., & Stillman, S. (2003). Instrumental variables and GMM: Estimation and
testing. The Stata Journal, 3(1), 1–31.
Bellemare, M. F., Masaki, T., & Pepinsky, T. B. (2017). Lagged explanatory variables and the estimation of
causal effect. Journal of Politics, 79(3), 949–963. https://​doi.​org/​10.​1086/​690946
Bellemare, M. F. (2014). Rise in food prices, food price volatility and policy troubles. American Journal of
Agricultural Econnomics, 96(5)
Berazneva, J., & Lee, D. R. (2013). Explaining the African food riots of 2007–2008: An empirical analysis.
Food Policy, 39, 28–39.
Berdal, M., & Malone, D. (2000). Greed and grievance: economic agendas in civil wars (p. 2000). Boulder:
Lynne Rienner Publishers.
Blundell, R., & Bond, S. (1998). Initial conditions and moment restrictions in dynamic panel data models.
Journal of Econometrics, 87(1), 115–143.
Bohlken, A. T., & Sergenti, E. J. (2010). Economic growth and ethnic violence: An empirical survey of
Hindu–Muslim communities in India. Journal of Peace Research, 47, 589–600.
Buhaug, H. (2010). The climate is not responsible for African civil war. PNAS, 107, 16477–16482.
Buhaug, H., Benjaminsen, T. A., Sjaastad, E., & Theisen, O. M. (2015). Climate variability, food production
shocks, and violent conflict in Sub-Saharan Africa. Environmental Research Letters, 10(12), 125015.
Burke, M., Miguel, E., Satyanath, S., Dykema, J., & Lobell, D. B. (2009). Rehabilitation increases the risk
of civil war in Africa. PNAS, 106, 20670–20674.
 A. B. Henri Aurélien, O. N. Bruno Emmanuel

Burke, B. M., Hsiang, S. M., & Miguel, E. (2015). Climate and conflict. Annual Tourism Economics, 7,
577–617.
Butkiewicz, J. L., & Yanikkaya, H. (2006). Institutional quality and economic growth: Maintenance of the
rule of law or democratic institutions, or both? Economic Modelling, 23(4), 648–661.
Caruso, R., Petrarca, I., & Ricciuti, R. (2014). Climate change, risk-taking and violence: Proofs of
Indonesia. Work Bouillie, 4665, CESifo, Munich
CCKP, Climate change Knowledge Portal (2021). The World Bank Group.
Collier, P., & Hoeffler, A. (2004). Greed and grievance in civil war. Oxford Economic Papers, 56(4),
563–595.
Dell, M., Jones, B. F., & Olken, B. A. (2012). Thermal shocks and economic growth: evidence of the last
half-century. Un m. Economics Journal Macroeconomics 4(3): 66–95 impacts, bonds and attenuation
measures. Rights and Resources Initiative, Washington DC
Demarest, L. (2015). Food price rises and political instability: Problematizing a complex relationship. The
European Journal of Development Research, 27, 650–671.
Detges, A. (2016). Local conditions of drought-related violence in sub-Saharan Africa: The role of road and
water infrastructures. Journal of Peace Research, 53(5), 696–710.
Devlin, C., & Hendrix, C. S. (2014). Trends and triggers redux: Climate change, rainfall, and interstate
conflict. Political Geography, 43, 27–39.
Dufumier, M., & Hugon, P. (2008). Piques Et Polémiques Les « Émeutes De La Faim » : Du Sous
Investissement Agricole À La Crise Sociopolitique. Revue Tiers Monde, 196(4), 927. https://​doi.​org/​
10.​3917/​rtm.​196.​0927
Dupasquier, C., & Osakwe, P. N. (2006). Foreign direct investment in Africa: Performance, challenges, and
responsibilities. Journal of Asian Economics, 17(2), 241–260.
Eberle, U., Rohner, D., & Thoenig, M. (2020). Heat and hate: climate security and farmer-herder conflicts in
Africa. ESOC Working Paper, 22, 79
Eggoh, J., & Kobbi, H. (2021). L’instabilité politique et ses déterminants: Un réexamen empirique. Revue
Francaise D’economie, 4, 219–263.
European University Institute. (2009). Overcoming Fragility in Africa. San Domenico di Fiesole: Robert
Schuman Centre for Advanced Studies.
Food and agriculture organization FAO. (2011). Global food situation: FAO indices of food prices. Rome,
Italy Https: //www.​fao.​org/​world​foods​ituat​ion/​wfs-​home/​foodp​rices​index/​en/> Accessed 21 Jun 2011
FAOSTAT. (2020). FAO statistics, food and agriculture organization of the united
Feeny, S., Posso, A., & Regan-Beasley, J. (2016). Handle with care: Fragile states and the determinants
of fragility. Applied Economics, 47(11), 1073–1085.
Fenske, J. (2010). The causal history of Africa: A response to Hopkins. Economic History of Developing
Regions, 25(2), 177–212.
Fosu, A. K. (2003). Political instability and export performance in sub-Saharan Africa. Journal of
Development Studies, 39(4), 68–83.
Gakpa, L. L. (2019). Instabilité politique, IDE et effets sur la croissance économique dans les
pays d’Afrique subsaharienne: un modèle à équations simultanées dynamiques. . Région et
Développement, (50)
Gnimassoun, B., & Santos, D. (2021). Robust structural determinants of public deficits in developing
countries. Applied Economics, 53(9), 1052–1076. https://​doi.​org/​10.​1080/​00036​846.​2020.​18240​
63
Gupta, S. (1991). "Stochastic models of interpurchase time with time-dependent covariates. Journal of
Marketing Research, 28(1), 1–15.
Hendrix, C., Haggard, S., & Magaloni, B. (2009). Grief and opportunity: food prices, political regime
and protest. New York: Conference document of the international studies association
Hendrix, C. S., & Salehyan, I. (2012). Climate change, rainfall, and social conflict in Africa. Journal of
Peace Research, 49(1), 35–50.
Hill, A. D., Johnson, S. G., Greco, L. M., O’Boyle, E. H., & Walter, S. L. (2021). Endogeneity: A review
and agenda for the methodology-practice divide affecting micro and macro research. Journal of
Management, 47(1), 10117.
Homer-Dixon, T. (1994). Environmental rarities and violent conflicts: Proof of case. International
Security, 19(1), 5–40.
Homer-Dixon, T. F. (1999). Environment, scarcity, and violence. Princeton University Press.
Hsiang, S. M., Burke, M., & Miguel, E. (2013). Quantification of climate influence on human conflicts.
Sciences, 341, 6151. https://​doi.​org/​10.​1126/​scien​ce.​12353​67
Hsiang, S. M., Burke, M., & Miguel, E. (2014). Convey the meta-analyzes of climatic conflicts: Response to
Buhaug et al. https://​doi.​org/​10.​1007/​s10584-​014-​1276-z
Climate change and political stability in African states: the…

Hugon, P. (2009). Le rôle des ressources naturelles dans les conflits armés africains. Hérodote, 134(3),
63–79.
Hugon, P., & Mayeyenda, A. (2003). The effects of price policies in the Coton-franc cotton zones: An
empirical analysis. Rural Economy, 275, 66–82.
ICRG, International Country Risk Guide (2018). The PRS Group. Political Risk Services., Information on:
https://​www.​prsgr​oup.​com/​Defau​lt.​aspx
Imai, K., & Kim, I. S. (2021). On the use of two-way fixed effects regression models for causal inference
with panel data. Political Analysis, 29(3), 405–415.
International Labour Office, I. (2017). Working towards sustainable development: Opportunities.
Intergovernmental panel on climate change (IPCC). (2014). Climate change 2014: impacts, adaptation,
and vulnerability. Part A: Global and sectoral aspects. Contribution of working group II to the fifth
assessment report of the intergovernmental panel on climate change. Cambridge: New York
Jensen, P. S., & Gleditsch, K. S. (2009). Pluie, growth and civil war: the importance of the place. Aix
Economics Defense, 20(5), 359–372.
Keho, Y. (2012). Le rôle des facteurs institutionnels dans le développement financier et écono-mique des
pays de l’UEMOA. Revue Economique Et Monétaire, 12, 9–43.
Koubi, V. (2019). Climate change and conflict. Annual Review of Political Science, 22, 343–360. https://​
doi.​org/​10.​1146/​annur​ev-​polis​ci-​050317-​070830
Koubi, V., Bernauer, T., Kalbhenn, A., & Spilker, G. (2012). Climate variability, economic growth, and
civil conflict. Journal of Peace Research, 49(1), 113–127.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (2019). The quality of government.
Journal of Law, Economics, and Organization, 15(1), 222–279.
Lagi, M., Bertrand, K.Z., & Bar-Yam, Y. (2011). The food crises and political instability in North Africa
and the Middle East. arXiv preprint arXiv:​1108.​2455
Lecoutere, E., D’Exelle, B., & Van Campenhout, B. (2010). Who engages in water scarcity conflicts? A
field experiment with irrigators in semi-arid Africa
Le Ster, M. (2011). Les liens entre conflits et environnement. Les Cahiers d’Outre-Mer. Revue de géographie
de Bordeaux, 64(255), 429–434.
Leitão, N. C., & Shahbaz, M. (2016). Economic growth, tourism arrivals and climate change. Bulletin of
Energy Economics (BEE), 4(1), 35–43.
Macauley, H., & Ramadjita, T. (2015). Les cultures céréalières : riz , maïs , millet , sorgho et blé :
Document de référence. Nourrir l’Afrique, Un Plan d’action Pour La Transformation de
l’agriculture Africaine, 21–23
Malthus, T. R. (1798). An essay on the principle of population. London: Johnson, reprinted, Oxford:
Oxford University Press, 1993.
Manarik, J.D. (1981). Modernization imbalances and political instability in black Africa 1960–1967.
Doctoral dissertation, Ohio University
Mares, D. M., & Moffett, K. W. (2016). Climate change and interpersonal violence: A “global” estimate
and regional inequities. Climatic Change, 135, 297–310.
Maystadt, J., & Ecker, O. (2014). Extreme meteorological conditions and civil war in Somalia: Does
security secure the conflict in Somalia due to the shocks of the price of beer? Un M Journal of
Agricultural Economics, 96, 1157–1182.
Maystadt, J., Calderone, M., & You, L. (2015). Local recovery and violent conflict in the North and
South Soudan. Journal of Economic Géography, 15, 649–671.
Minot, N. (2014). Food price volatility in sub-Saharan Africa: Has it really increased? Food Policy, 45,
45–56.
N’Dimina-Mougala, A. D. (2007). Les conflits Africains au XXe siècle. Essai de typologie. Guerres
Mondiales Et Conflicts Contemporains, 225(1), 121–131. https://​doi.​org/​10.​3917/​gmcc.​225.​0121
Nelson, G.C., Rosegrant, M.W., Koo, J., Robertson, R., Sulser, T., Zhu, T., & Lee, D. (2009). Climate
change: Impact on agriculture and costs of adaptation. Intl Food Policy Res Inst. 21
O’Loughlin, J., Witmar, F., Linke, A., Lainge, A., Gettleman, A., & Dudhia, J. (2012). Climatic
variability and conflict risk in East Africa, 1990–2009. Proceedings of the National Academy
Science United States, 109(45), 18344–18349.
ONU, W. (2020). Rapport mondial des Nations Unies sur la mise en valeur des ressources en eau 2020
. Paris, 2020, disponible sur : https://​www.​un.​org/​en/​events/​deser​tific​ation​day/​deser​tific​ation.
shtml.: L’eau et les changements climatiques, UNESCO
Parry, M. L. (2009). Assessing the costs of adaptation to climate change: A review of the UNFCCC and
other recent estimates
Radu, M. (2015). Political stability-a condition for sustainable growth in Romania? Procedia Economics
and Finance, 30, 751–757.
 A. B. Henri Aurélien, O. N. Bruno Emmanuel

Raleigh, C., & Kniveton, D. (2012). Whether it is raining or selling: An analysis of conflicts and climatic
variability. Journal of Peace Research, 49(1), 51–64.
Raleigh, C., & Urdal, H. (2007). Climate change, environmental degradation and military conflict.
Politics Géography, 26(6), 674–694.
Raleigh, C., Choi, H. J., & Dominic, K. (2015). The devil is in the details: an investigation into
the relations between the conflicts, the food prices and the climate through Africa. Global
Environmental Change, 32, 187–199.
Roodman, D. (2009). How to do xtabond2: An introduction to difference and system GMM in Stata The
stata journal. Stata Journal, 9, 86–136.
Sarsons, H. (2011). Precipitations and conflicts. Cambridge, MA: Harvard University,.
Sofuoğlu, E., & Ay, A. (2020). The relationship between climate change and political instability: The
case of MENA countries (1985: 01–2016: 12). Environmental Science and Pollution Research,
27(12), 14033–14043.
Theisen, O. M. (2012). Climate clashes? Weather variability, land pressure, and organized violence in
Kenya, 1989–2004. Journal of Peace Research, 49(1), 81–96.
Von Uexkull, N. (2014). Security extended, vulnerability and civil conflicts in sub-Saharan Africa.
Géography Politique, 43, 16–26.
Uexkull, V., N, Croicu, M., Fjelde, H., & Buhaug, H. (2016). Civil conflict sensitivity to growing-season
drought. Proceedings of the National Academy of Sciences, 113(44), 12391–12396.
UN, United Nations (2020). Climate change annual reports. United Nations.
UN Security Council. (2009). Report of the International Commission of Inquiry mandated to establish the
facts and circumstances of the events of 28 September 2009. S/2009/693,66259(18 December 2009),
Guinea.: United Nations
UN Security Council, U. (2019). Climate Change Recognised as Threat Multiplier. UN Security Council
Debates Its Impact on Peace. (Geneva: United Nations); at: https://​news.​un.​org/​en/​story/​2019/​01/​
10313​22: United Nations.
Weezel, S. (2020). Local warming and violent armed conflict in Africa. World Development, 126,
104708. https://​doi.​org/​10.​1016/j.​world​dev.​2019.​104708
Women, U.N. (2022). Progress on the sustainable development goals: The gender snapshot 2022
Wood, R. M., & Wright, T. M. (2016). Responding to catastrophe: Repression dynamics following rapid-
onset natural disasters. Journal of Conflict Resolution, 60(8), 1446–1472.
Wooldridge, J.M. (2021). Two-way fixed effects, the two-way mundlak regression, and difference-in-
differences estimators. Available at SSRN 3906345
WGI, World Governance Indicator. (2021). Update https://​www.​govin​dicat​ors.​org
WMO, World Meteorological Organization. (2020). State of the climate in Africa 2019. Annex VII of the
WMO Meteorological Technical Regulations. 2019 edition.
World, B. (2013). Political stability and absence of violence / terrorism. 1–2. http://​info.​world​bank.​org/​
gover​nance/​wgi/​Home/​Docum​ents#:​~:​text=​Polit​ical Stability and Absence of Violence/Terrorism
Xu, T. (2011). The determinants of political instability: a regression analysis. Thèse de doctorat.
Marietta College
Yabré, T., Semedo, G., & Ouédraogo, I. M. (2020). Stabilité du gouvernement et ajustement budgétaire de
pays d’Afrique subsaharienne. Revue Française D’économie, 34(4), 25–69. https://​doi.​org/​10.​3917/​
rfe.​194.​0025
Yeyouomo, A. K., & Asongu, S. (2023). Sustainable urbanization and vulnerability to climate change in
Africa: Accounting for digitalization and institutional quality. Sustainable Development, 3, 1–29.
https://​doi.​org/​10.​1002/​sd.​2721
Zhang, Z., Lohr, L., & C, E., & Wetzstein, M. (2010). Nutrition against fuel: What do we know about
prices? Food Policy, 38(1), 445–451.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under
a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted
manuscript version of this article is solely governed by the terms of such publishing agreement and applicable
law.