Food Security and Weather Events: A Multidimensional Analysis in the West African Sahel for 2001

Abstract

This paper analyzes the climatic factors that affected food security in the West African Sahel in 2001–2017. We estimate the impact of droughts and floods on the four dimensions of food security defined by the Food and Agricultural Organization of the United Nations, based on a panel data model controlling for socioeconomic and political factors. Droughts and floods negatively affect food security with floods causing more damage. Since socioeconomic and political factors, especially conflicts, also play an important role, food security in the West African Sahel cannot be explained only by climatic problems, so that coordinated policies must be based on the four dimensions of food security.

Keywords

Food security droughts floods weather events West African Sahel

Introduction

Sustainable agrifood systems should provide food and nutrition security. Food security was first mentioned in Rome in 1976 at the World Food Conference, and, since then, its definition and measurement have considerably evolved (Ashley, 2016; Maxwell, 1996) and have even changed (Jones et al., 2013). There are now almost 200 definitions of this concept (Ashley, 2016) and a multitude of indicators exist together with several attempts for constructing composite indicators. We retain the most shared definition after the World Food Summit of November 1996 (Ashley, 2016): “Food security exists when all people, at all times, have physical and economic access to sufficient, safe and nutritious food that meets their dietary needs and dietary preferences for an active and healthy lifestyle.” Conversely, food insecurity results from the fact that people do not have adequate physical, social or economic access to food. In this context, climate change obviously has multiple direct and indirect effects on food security by affecting growth and income distribution, offer and demand for agricultural products, and so on (Akpaki et al., 2020; Douglas, 2017).

More precisely, food security has four components: food availability, accessibility, use and stability of food (Ashley, 2016; FAO, 2014). The first dimension pertains to availability. Food may be available in the family farm, attic, kitchen, store and local market (Ashley, 2016; Reig, 2012). At the country or regional level, food supply can come from domestic production, imports, stocks, and food aid (Reig, 2012). The second dimension raises an essential problem linked to inequalities (poverty), especially in countries where infrastructure is not developed (Ashley, 2016). Food security must take into account ease of access of people to food. The third dimension is linked to the effects of eating the food and to better use of food, water quality, and access to care for individuals. Finally, the fourth dimension refers to the interaction of these three dimensions and determines the food security status and stability of a country or a household over time.

In this paper, we focus on food security in the West African Sahel. While more than 113 million people in 53 countries are severely food insecure, according to the Food Security Information Network (FSIN, 2019) in its 2019 Global Food Crisis Report, the situation in the Sahel countries has received special attention as it is further exacerbated by terrorism. CILSS (2019) points out that although food availability may be satisfactory, civil insecurity is the main cause of the destruction of livelihoods and severe food insecurity in the Lake Chad basin and north of Mali. More generally, conflicts do not allow the mobilization and transport of food and livestock in certain areas or countries. Consequently, although these countries have domestic production, they must use imports to meet domestic needs. This mismatch between the drop in domestic production and domestic demand poses a major problem for the Sahelian economies.

Other factors obviously affect food security in the Sahel. Rising commodity prices are often seen on local and international markets. Climate change and the associated increased probability of extreme weather events affects agricultural productivity and 80% of the Sahelian agricultural population (Davis et al., 2010; Yobom and Le Gallo, 2022) is at risk of being food insecure, following an increase in the occurrence of drought or flooding. This can limit the purchasing power of poor households and push these individuals to focus on cheaper grains and other less nutritious foods. In recent decades, widespread inflation has become one of the main sources of migration from urban populations to rural areas. Growing rural population also increases density, which is likely to lead to food insecurity. The opposite effect is also observable and has become more significant with the influx of migrants toward cities, which also creates a significant food imbalance in the recipient areas. These phenomena cause significant human loss and economic damage in the Sahel countries due to the lack of good quality infrastructure. The Covid-19 pandemic has made the situation even more difficult. Despite the diversity of countries and food systems, the countries of the Sahel globally face an exposure of their flaws in a context where this pandemic has been a significant test of the resilience capacities of their economic, food, political, and social systems.

In this paper, our aim is then to assess the impact of extreme weather events on food security by controlling for several socioeconomic and political factors. We use a fixed effects panel model estimated on a sample of six countries in the West African Sahel (Burkina Faso, Mali, Mauritania, Niger, Nigeria, and Senegal) for the period 2001–2017. Fixed effects allow controlling for time invariant unobserved heterogeneity (Blanc and Schlenker, 2017). Our specification includes lagged weather variables to mitigate the problem of simultaneity. The fixed effects specification is well suited to assess the impacts of weather events, because it uses group fixed effects to absorb all the invariant variations over time. In addition, by building food security indices and using climate shock variables such as droughts and floods as in Kinda and Kere (2016) and Karfakis et al. (2011), the paper contributes to the literature in several ways by focusing on the Sahel countries.

First, this paper considers all four dimensions of FAO to build composite food security indices to better describe each dimension of food security. This contrasts with Belloumi (2014); Reyes et al. (2014); Asfaw (2015); Ahmad Munir and Iqbal (2016); Kinda and Kere (2016) who use proxies (i.e. food production for Belloumi, 2014; vulnerability index for Karfakis et al., 2011; crop net income for Asfaw, 2015; value of food consumed by adult equivalent for Reyes et al., 2014; etc.) to directly assess food security. To our knowledge, no study has empirically assessed food security by adopting a multidimensional analysis in the Sahel countries. Some studies have studied this link in East African countries (Belloumi, 2014), in households in Ethiopia (Asfaw, 2015), in South Africa (Masipa, 2017) and in 53 developing countries with unidimensional indices (Kinda and Kere, 2016). Next, this paper contributes to the literature on the link between weather events and food security in the Sahel countries by showing the importance of climatic shocks (floods and droughts) of the previous year on all dimensions of food security. Also, the paper examines the important role of other socioeconomic and political factors on food security in the Sahel. Finally, we assess this link for a newly defined Sahel taking into account all the countries from Senegal to the Horn of Africa.

The remainder of the paper is organized as follows. The second section briefly presents a literature review. The third section presents the data used and the methodology adopted to construct the multidimensional index of food insecurity. The fourth section presents the model estimated. The estimation results and interpretations are presented in the fifth section. Finally, conclusion and implications are presented in the sixth section.

Literature review

In the empirical literature, various approaches have been used to assess the link between climate change/weather events and food security. These studies can be distinguished by the definition and construction of the food security variable and by the type of identification strategy used, that is, linear regression models with ordinary least squares or instrumental variables (Asfaw, 2015; Ervin and Gayoso de Ervin, 2019; Karfakis et al., 2011; Reyes et al., 2014). Some authors use proxies (Belloumi, 2014; Kinda and Kere, 2016) while others construct a food security index (Ahmad Munir and Iqbal, 2016). Table A2 in the appendix summarizes the articles that construct food security indicators and analyze the link between climate change/weather events and food security. Most studies focus on the poorest and therefore most vulnerable countries where food insecurity is a major problem for populations. Starting with Kinda and Kere (2016), developing countries have been widely studied with a focus on South American (Ervin and Gayoso de Ervin, 2019; Karfakis et al., 2011; Reyes et al., 2014), Asia (Ahmad Munir and Iqbal, 2016) and African (Asfaw, 2015; Belloumi, 2014; Masipa, 2017) countries.

The spatial scales of analysis and units of observations are diverse: agricultural producers (Ahmad Munir and Iqbal, 2016), households (Asfaw, 2015; Ervin and Gayoso de Ervin, 2019; Karfakis et al., 2011), provinces (Reyes et al., 2014), or countries (Belloumi, 2014; Kinda and Kere, 2016; Masipa, 2017).

The variables explained are also different. Some authors (Ahmad Munir and Iqbal, 2016; Asfaw, 2015; Belloumi, 2014; Kinda and Kere, 2016; Reyes et al., 2014) use proxies directly (i.e. food production for Belloumi, 2014; vulnerability index for Karfakis et al., 2011; net income of crops for Asfaw, 2015; value of food consumed by the adult equivalent for Reyes et al., 2014 and Kinda and Kere, 2016; combinations of proxies (agricultural productivity, calories consumption per capita, food consumption) for Ervin and Gayoso de Ervin, 2019), and others such as Ahmad Munir and Iqbal (2016) construct a food security index.

The models include climate and/or weather variables together with socioeconomic characteristics. Climate and weather variables generally refer to temperature and precipitation (Ahmad Munir and Iqbal, 2016; Asfaw, 2015; Belloumi, 2014; Ervin and Gayoso de Ervin, 2019; Karfakis et al., 2011; Masipa, 2017; Reyes et al., 2014), or climatic shocks such as droughts, floods, and extreme temperatures (Kinda and Kere, 2016).

Using binary models, linear regression with instrumental variable or panel data, the effects of climatic and weather variables vary according to scales but remain almost unanimous: climate change and weather events affect food security. Climate disrupts agricultural production, reduces food availability and also affects the distribution of food. For example, Kinda and Kere (2016) show that climate change, measured by variability in the water balance, droughts, floods, and extreme temperatures, reduces food availability in the affected countries. Ervin and Gayoso de Ervin (2019) show that increasing temperatures and decreasing rainfall reduce agricultural productivity and caloric consumption, and increase vulnerability to food insecurity. Masipa (2017) shows that climate change poses a high risk for food security in sub-Saharan countries, from agricultural production to food distribution and consumption.

However, most papers do not consider the fact that food security is a multidimensional dimension. Only one proxy or variable may not provide a relevant measure encompassing all dimensions of this concept. In this paper, we set up a multidimensional approach and we build our dimensional indices on the basis of several components and indices, available on the website of FAO (https://www.fao.org/faostat/fr/#home). After having presented the study area, we turn to the construction of our indices below.

Data

Study area

Our study area covers 6 countries of the West African Sahel (Burkina Faso, Mali, Mauritania, Niger, Nigeria, and Senegal, see Figure 1), for the period 2001–2017. These six countries cover a Sahelian zone that includes all arid and semi-arid countries with the same climatic characteristics (variable and hot climate, uncertain rainy seasons, occurrence of regular floods and droughts), bordered by desert and facing similar sensitivity to climate shocks generating famines and food insecurity.

Figure 1.

Study area.

In Sahelian Africa, about 90% of the population depends on subsistence agriculture (Sterk, https://www.fao.org/faostat/fr/#home). Agriculture, employing about 80% of the active population, and livestock remain the main sources of economic wealth. Rainfed agriculture remains dominant and is practiced only during the 3 months of the rainy season (Yobom, 2020). The absence of an irrigated agricultural system then diminishes wealth creation, as farmers are inactive the rest of the year, and it also makes rural population the most impacted by climate change and variability.

In addition, the Sahelian countries display the lowest rate of industrialization compared with the rest of the world, with the exception of Nigeria. They also have the lowest rate of economic diversification in the world. Moreover, these countries are confronted with civil wars, massive migrations (internal or external) and terrorism, which create lasting instability. Facing all these difficulties, the fields are deserted, the farmers leave and take refuge far from their fields.

Consequently, the countries of the Sahel are among the poorest countries in the world. At the same time, they are characterized by very strong population growth (around 3.1%) and rapid urbanization rate estimated at around 7%. All these issues worsen the problem of food self-sufficiency making this area very sensitive and fragile. In this context, policies aimed at reducing famine and food insecurity are organized through a combination of efforts between local governments and their various partners, bilateral (agreements between countries) and multilateral (nongovernmental organizations and international institutions).

Food security indicators

Food security indicators must capture the four dimensions of availability, access, use, food stability over time or a combination of these determinants (Babu et al., 2014; Jones et al., 2013). However, it is too restrictive to capture each food security dimension with just one variable. We therefore develop separate indices for each of the dimensions of food security. For that purpose, we use principal component analysis (PCA) (Ahmad Munir and Iqbal, 2016; Husson et al., 2016), which allows the construction of linear combinations of a set of correlated indicators, reducing them into factors while extracting the greatest variance from the original variables.

The variables on which we base PCA come from the definition of the FAO which provides a list of food security indicators for each of the four dimensions. From them, we retained the indicators that are available for our study area over the period. Table A1 in the appendix provides detailed information for the variables we retain. Moreover, most variables are provided as an average of 3 years in the FAO database. Therefore, for consistency, we constructed all variables similarly.

These indicators are then used as a basis for the four PCAs to build the four food security indices (availability index, accessibility index, use index, and stability index), the results of which are displayed in Figures A1 to A4 in the appendix. The objective is to capture the weights of the variables on the first dimension of the PCA that are used as the food security indices. The red dotted line on the different graphs in Figures A1 to A4 gives the expected average contribution. If the contribution of the variables was uniform, the expected value would be 1/length(variables) = 1/10 = 10% (Kassambara, 2017). We provide further details for each food security dimension below.

Availability

This first dimension captures the availability of food and includes all types of food regardless of their source (domestic production, imports, stocks and food aid). The set of indicators included to capture this dimension reflects the supply of food in terms of energy, per capita value and supply. PCA is applied to 5 indicators for this first dimension and their descriptive statistics are given in Table 1. They are:

Average dietary energy supply (DES) adequacy (percent) (3-year average). The indicator expresses the DES as a percentage of the average dietary energy requirement (ADER). Each country’s or region’s average supply of calories for food consumption is normalized by the ADER estimated for its population to provide an index of adequacy of the food supply in terms of calories.

DES used in the estimation of prevalence of undernourishment (kcal/cap/day) (3-year average)

Share of DES derived from cereals, roots and tubers (kcal/cap/day) (3-year average). This indicator is expressed as a percentage of the total DES (in kcal/caput/day) and captures the diversity of food supply at the country level. In other words, it indicates the range of consumption choices and nutrients available to the population.

Average protein supply (g/cap/day) (3-year average)

This indicator measures the average national protein supply (expressed in grams per capita and per day) and provides information on the quality of the diet of the population of the countries.

Average supply of protein of animal origin (g/cap/day) (3-year average) The indicator measures the average national protein supply (expressed in grams per capita per day, e.g. meat, seafood, fish) and also provides details on the quality of the diet.

Table 1.

Descriptive statistics of the indicators for the first dimension of food security.

Dimension	Indicator	Description	Source	Mean	Std. dev.
Availability	var1	Average dietary energy supply adequacy (percent) (3-year average)	FAO	120.83	8.868
	var2	Average value of food production (constant 2004–2006 I$/cap) (3-year average)	FAO	150.45	55.84
	var3	Dietary energy supply used in the estimation of prevalence of undernourishment (kcal/cap/day) (3-year average)	FAO	2612.74	184.05
	var4	Share of dietary energy supply derived from cereals, roots and tubers (kcal/cap/day) (3-year average)	FAO	62.31	5.20
	var5	Average protein supply (g/cap/day) (3-year average)	FAO	72.02	9.30
	var6	Average supply of protein of animal origin (g/cap/day) (3-year average)	FAO	16.85	7.51
Countries	6
Sample size	103

FAO: Food and Agricultural Organization.

The results of the construction of the first dimension are grouped together in the graphic series in the appendix (Figure A1 and Table A3 in the appendix). Axis 1 has an inertia of 60.40%. In addition, the correlation circle in Supplemental Figure A1 shows that all variables are correlated to axis 1 (except var3) and clearly contribute to its construction: var1 (23.29%), var2 (25.39%), var3 (12.30%), var4 (20.61%), and var5 (18.39%). The variables var1, var2 and var5 have a very high and positive correlation close to 1 indicating that they are strongly related to the coordinates on axis 1. Finally, Supplemental Figure A1 and Table A3 show that these variables are also very well projected because they are close to the correlation circle. These contributions are used to build the availability index called “Ind_availability.”

Accessibility

This second dimension captures all aspects related to the social, physical and economic access of populations to available food. Due to data unavailability, we only have three variables for this second dimension for the countries in our sample and time period (Table 2).¹

Rail lines density (total route in km per 100 square km of land area). Rail lines density corresponds to the ratio between the length of railway route available for train service, irrespective of the number of parallel tracks (rail lines, total route in km) with the area of the country.

Prevalence of undernourishment (percent) (3-year average). This indicator is expressed in terms of the proportion of undernourished people in the population of a given country or region. Furthermore, the measure is not based on the actual count of undernourished people in a country, but on the probability that a person is undernourished.

Number of people undernourished (million) (3-year average). This indicator measures the number of undernourished people in each country. It brings together all the people suffering from malnutrition and who are in a situation of undernourishment.

Table 2.

Descriptive statistics of the indicators for the second dimension of food security.

Dimension	Indicator	Description	Source	Mean	Std. dev.
Accessibility	var1	Rail lines density (total route in km per 100 square km of land area)	FAO	0.22	0.18
	var2	Prevalence of undernourishment (percent) (3-year average)	FAO	11.55	4.97
	var3	Number of people undernourished (million) (3-year average)	FAO	3.80	5.20
Countries	6
Sample size	103

FAO: Food and Agricultural Organization.

The results of the construction of the second dimension are given in Figure A2 and Table A4 in the appendix. The results indicate that the variables do not have the same inertias and contribute in a distinct way to the construction of axis 1. The difference is explained by the first dimension, that is, 47.59%, using the elbow criterion which allows the axes to be selected before the offset. In addition, the variables are strongly projected due to their proximity to the circle of correlations. The factors var1, var3 and var2 respectively contribute 45.44%, 44.24%, and 10.31% to the construction of the first dimensions. These contributions are used to build the accessibility index called “Ind_accessibility.”

Utilization

This dimension relates to the use of food and is concerned with the use people make of available food in terms of nutritional value. It provides an overview of the effect of food on consumers while allowing the assessment of the health status of consumers (Ashley, 2016; FAO, 2014). Table 3 gives the descriptive statistics of the 5 indicators used in the PCA of this third dimension.

Table 3.

Descriptive statistics of the indicators for the third dimension of food security.

Dimension	Indicator	Description	Source	Mean	Std. dev.
Utilization	var1	Percentage of population using at least basic drinking water services	FAO	57,66	11.15
	var2	Percentage of population using at least basic sanitation services (percent)	FAO	27.68	13.18
	var3	Percentage of children under 5 years of age who are stunted (modeled estimates) (percent)	FAO	34.91	9.38
	var4	Percentage of children under 5 years of age who are overweight (modeled estimates) (percent)	FAO	2.29	1.05
	var5	Prevalence of anemia among women of reproductive age (15–49 years)	FAO	53.83	4.82
Countries	6
Sample size	103

FAO: Food and Agricultural Organization.

Percentage of population using at least basic drinking water services. Percentage of people using basic water services as well as those using safely managed water services. Basic drinking water services are defined as drinking water from an improved source, provided collection time is not more than 30 minutes for a round trip. Improved water sources include piped water, boreholes or tubewells, protected dug wells, protected springs, and packaged or delivered water.

Percentage of population using at least basic sanitation services. Percentage of people using at least basic sanitation services, that is, improved sanitation facilities that are not shared with other households. This indicator encompasses both people using basic sanitation services as well as those using safely managed sanitation services. Improved sanitation facilities include flush/pour flush to piped sewer systems, septic tanks or pit latrines; ventilated improved pit latrines, compositing toilets or pit latrines with slabs.

Percentage of children under 5 years of age who are stunted (modeled estimates). Percentage of stunting (height-for-age less than −2 standard deviations of the WHO Child Growth Standards median) among children aged 0–5 years.

Percentage of children under 5 years of age who are overweight (modeled estimates). Percentage of overweight (weight-for-height more than 2 standard deviations of the WHO Child Growth Standards median) among children aged 0–5 years.

Prevalence of anemia among women of reproductive age (15–49 years). Prevalence of anemia among women of reproductive age refers to the combined prevalence of both non-pregnant with hemoglobin levels below 12 g/dL and pregnant women with hemoglobin levels below 11 g/dL.

The results of the construction of the third dimension are given in Figure A3 and Table A5 in the appendix. Axis 1 groups together 55.69% inertia, thus explaining most of the variability. The factors var1, var2, var3, var4 and var5 contribute respectively to 31.60%, 32.32%, 29.68%, 0.538%, and 5.84% in the construction of dimension 1. Var2, var1 and var5 are positively correlated with axis 1, close to 1. In addition, they are also well projected because they are close to the circle of correlations while being very related to the coordinates on axis 1, leading thus the construction of the dimension “Ind_utilization.”

Stability

At the country level, food stability is achieved when the entire population receives food at all times. It concerns the stability and sustainability of the food source over time (Ashley, 2016; Aurino, 2014; FAO, 2014). Four indicators are used in the PCA for this fourth dimension, the descriptive statistics of which are displayed in Table 4.

Percentage of arable land equipped for irrigation (3-year average) Ratio between arable land equipped for irrigation and total arable land. Arable land is defined as the land under temporary agricultural crops (multiple-cropped areas are counted only once), temporary meadows for mowing or pasture, land under market and kitchen gardens and land temporarily fallow (less than 5 years). The abandoned land resulting from shifting cultivation is not included in this category. Data for arable land are not meant to indicate the amount of land that is potentially cultivable. Total arable land equipped for irrigation is defined as the area equipped to provide water (via irrigation) to the crops. It includes areas equipped for full and partial control irrigation, equipped lowland areas, pastures, and areas equipped for spate irrigation.

Value of food imports in total merchandise exports (percent) (3-year average): Value of food (excl. fish) imports over total merchandise exports.

Per capita food production variability (constant 2004–2006 thousand int$ per capita): Per capita food production variability corresponds to the variability of the “food net per capita production value in constant 2004–2006 international $” as disseminated in FAOSTAT.

Per capita food supply variability (kcal/cap/day): Per capita food supply variability corresponds to the variability of the “food supply in kcal/caput/day” as disseminated in FAOSTAT.

Table 4.

Descriptive statistics of the indicators for the fourth dimension of food security.

Dimension	Indicator	Description	Source	Mean	Std. dev.
Stability	var1	Percent of arable land equipped for irrigation (3-year average)	FAO	3.78	3.77
	var2	Value of food imports in total merchandise exports (percent) (3-year average)	FAO	31.24	19.55
	var3	Per capita food production variability (constant 2004–2006 thousand int$ per capita)	FAO	10.20	4.65
	var4	Per capita food supply variability (kcal/cap/day)	FAO	36.80	15.56
Countries	6
Sample size	103

FAO: Food and Agricultural Organization.

The results of the construction of the fourth dimension are given by Figure A4 and Table A6 in the appendix. Axis 1 represents 40.6% inertia and axis 2 approximately 30%. The factors var1, var2, var3 and var4 respectively contribute 22.351%, 24,60%, 51.92%, and 1.11% in the construction of axis 1. We have a very strong correlation for the two variables close to 1, which means that var2 and var3 are very related to the coordinates on axis 1. Moreover, Supplemental Figure A4 and Table A6 show that these variables are well projected as they are close to the circle of correlations. These contributions are sufficient to build the stability index called “Ind_stability.”

Summarizing these results, in the four PCAs, the first component explains respectively 51.5%, 47.6%, 55.7%, and 40.6% of the total variance of the variables. These figures are globally higher or similar to the ones in the study by Bilan et al. (2018) who used a factor with an inertia of 48.37% to assess the impact of environmental determinants on the state of food security in 28 post-socialist countries, those in the study by Reig (2012) who studied food security in African and Arab countries and obtained a proportion of 56.2%, those of Demeke et al. (2011) who used a factor explaining 32.5% of the total variance to study the effects of climate change on food security in Ethiopia, and those of Nyaga and Doppler (2009) extracted a proportion of 34% to predict the factors affecting food security in the district of Murang’a in Kenya.

Variables of interest

We measure climate change by the occurrence of extreme events such as droughts and floods as in Kinda and Kere (2016). These two events are natural disasters provided by the database (EM-DAT) of the Center for Research on the Epidemiology of Disasters (CRED). The definitions of droughts and floods given by CRED are presented in Table 5 together with descriptive statistics. Notably, drought is an abnormal deficit over a prolonged period. We include the number of events that occurred in a country in a given year, because these two shocks are distinguished by their intensity and the damage (human loss, economic and monetary damage) caused in the countries. The inclusion of these two variables of climatic interest makes it possible to verify their direct links with the food security situation in the Sahel countries with expected heterogenous results along the four dimensions. Figure 2 shows the spatial distribution of droughts and floods in the Sahel countries from 2001 to 2017.

Table 5.

Descriptive statistics for the variables of interest, drought, and flood.

Variables	Description	Source	Mean	Std. dev.
Drought	An extended period of unusually low precipitation that produces a shortage of water for people, animals and plants. Drought is different from most other hazards in that it develops slowly, sometimes even years, and its onset is generally difficult to detect. Drought is not solely a physical phenomenon because its impacts can be exacerbated by human activities and water supply demands. Drought is therefore often defined both conceptually and operationally. Operational definitions of drought, meaning the degree of precipitation reduction that constitutes a drought, vary with locality, climate and sector.	EM-DAT	0.19	0.42
Flood	A general term for the overflow of water from a stream channel onto normally dry land in the floodpain (riverine flooding), higher than normal levels along the coast and in lakes or reservoirs (coastal flooding) as well as ponding of water at or near the point where the rain fell (flash floods)	EM-DAT	1	0.89
Countries	6
Sample size	103

Figure 2.

Total occurrence of the two natural climatic disasters from 2001 to 2017 in the Sahel countries. (a) Drought; (b) flood.

Socioeconomic and political variables

The set of control variables is composed of agricultural factors (cereal yield), level of economic development (measured by real gross domestic product (GDP) per capita), demographic characteristics of population (demographic growth), imports (cereal imports), political stability and the absence of violence (stability), and inflation (consumer prices). The control variables selected are presented in Table 6, together with descriptive statistics, and described below.

Cereal yield. Cereal yield proxies the cereal production of the countries. Cereal yield can contribute to food security as it can inform about the availability of cereals at country level. This variable thus makes it possible to control the impact of agricultural production on food security.

GDP. The inclusion of GDP per capita helps controlling the impact of wealth of the population on food security. It plays an important but indirect role in improving food security. Indeed, it allows populations to access available food, and it can also have effects on all the other dimensions.

Population growth. Growing population leads to increased demand for food. Faced with limited food resources (water resources, agricultural land and infrastructure), population growth reduces these resources and constitutes an obstacle to the achievement of food security. We therefore expect that higher population growth will lead to poor food security.

Cereal imports. Cereal imports are intended to help meeting the domestic food deficit. Imports should improve domestic food availability.

Stability. Stability is a very important aspect for guaranteeing food security. While political and social stability can guarantee the achievement of food security, instability linked to terrorism and armed/civil conflicts can disrupt the food chain from the field to marketing.

Inflation. Inflation affects people’s food purchasing decisions and can also affect producers. Inflation has negative effects on food security by reducing food availability, making food inaccessible to people.

Migration. Mass migration as a result of insecurity issues or civil wars can put strong pressure on food demand in host countries.

Table 6.

Descriptive statistics for the control variables.

Variables	Description	Source	Mean	Std. dev.
GDP per capita	GDP at market prices is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the product (US$)	WB	1008.00	651.40
Population growth	Annual population growth rate. Population is based on the de facto definition of population, which counts all residents regardless of legal status or citizenship.	FAO	2.99	0.40
Cereal yield	Cereal yield measured as kilograms per hectare of harvester land, includes wheat, rice, maize, barley, oats, rye, mil- let, sorghum, buckwheat, and mixed grains. Production data on cereals relate to crops harvested for dry grain only. Cereal crops harvested.	FAO	1050.41	365.14
Inflation price	Inflation as measured by the consumer price index reflects the annual percentage change in the cost to the average consumer of acquiring a basket of goods and services that may be fixed or changed at specified intervals, such as yearly. The Laspeyres formula is generally used.	FAO	4.33	4.86
Cereal imported	Imports data (in metric tons) on cereals relate to crops harvested for dry grain only. Cereal crops harvested green for food, feed, or silage and those used for grazing are excluded. (1000 metric tons)	FAO	1387.95	1817.79
Political Stability	Political Stability and Absence of Violence/Terrorism measures perceptions of the likelihood of political instability and/or politically-motivated violence, including terrorism. Estimate gives the country’s score on the aggregate indicator, in units of a standard normal distribution, i.e. ranging from approximately −2.5 to 2.5.	FAO	−0.75	0.77
Migration	Net migration is the net average annual number of mi- grants during the period, that is, the annual number of immigrants less the annual number of emigrants, including both citizens and non-citizens. Data are 5-year estimates.	Afdb	−0.616	1.30
Countries	6
Sample size	103

GDP: gross domestic product; FAO: Food and Agricultural Organization; WB: World Bank (available https://data.worldbank.org/indicator); Afdb: African Developmet Bank (available https://dataportal.opendataforafrica.org).

Model

We use panel data models as they allow to have an overview of the phenomenon over several years and lag our variable of interest: past climate shocks affect present consumption. Simultaneity problems are also mitigated. In addition, food security in the Sahel countries depends on other socioeconomic and political factors that we include in the model.

Our benchmark model (1) includes only the control variables:

\begin{array}{l} F S_{k, i, t} = α_{i} + β_{1} Y i e l d_{i, t - 2} + β_{2} G D P_{i, t - 2} + β_{3} P o p_{i, t - 2} + β_{4} I m p o r t_{i, t - 2} \\ + β_{5} S t a b i l i t y_{i, t - 2} + β_{6} I n f l a t i o n_{i, t - 2} + β_{7} M i g r a t i o n_{i, t - 2} + ϵ_{i, t} \end{array}

(1)

where FS_{k, i,t} is the food security indicator k of country i in year t. α_i is the country specific constant if the model is estimated by a fixed effects model and is part of the error term if estimated by random effects model. Yield_i,t is the cereal yield of country i for year t. GDP_i,t is the Gross Domestic Product per capita of country i for year t. Pop_i,t denotes the population growth of country i in year t. Import_i,t and Stability_i,t respectively represent cereal imports and the instability and violence in country i during year t. Finally, Inflation_i,t and Migration_i,t denote respectively the inflation of the prices of consumer products and the net average annual number of migrants for the country i during the year t. The explanatory variables are lagged twice since, as mentioned in the fourth section, all food security variables are provided for each year t, as an average of values for t−1, t and t + 1. Institutional variables are very stable in time, so that we consider that their effects are captured by the individual effects. The total sample size is 103 observations.

We then include sequentially each weather event in model (2):

\begin{array}{l} F S_{k, i, t} = α_{i} + β_{1} S h o c k_{i, t - 2} + β_{1} Y i e l d_{i, t - 2} + β_{2} G D P_{i, t - 2} + β_{3} P o p_{i, t - 2} + β_{4} I m p o r t_{i, t - 2} \\ + β_{5} S t a b i l i t y_{i, t - 2} + β_{6} I n f l a t i o n_{i, t - 2} + β_{7} M i g r a t i o n_{i, t - 2} + ϵ_{i, t} \end{array}

(2)

where Shock_i,t₋₂ is the occurrence of drought or flood for country i if there is a shock in year t −2.

Finally, we also estimate the following final model (3) with all variables included:

\begin{array}{l} F S_{k, i, t} = α_{i} + β_{1} D r o u g h t_{i, t - 2} + β_{2} F l o o d_{i, t - 2} + β_{3} Y i e l d_{i, t - 2} + β_{4} G D P_{i, t - 2} + β_{5} P o p_{i, t - 2} \\ + β_{6} I m p o r t_{i, t - 2} + β_{7} S t a b i l i t y_{i, t - 2} + β_{8} I n f l a t i o n_{i, t - 2} + β_{7} M i g r a t i o n_{i, t - 2} + ϵ_{i, t} \end{array}

(3)

where Drought_i,t₋₂ and Flood_i,t₋₂ are the occurrence of these two climatic shocks of country i in year t −2. These events are independent and their occurrence differs from one country to another. Finally, note that the Hausman test shows that the fixed-effect panel model is more appropriate in all specifications.

Results and discussion

In all tables, the first column presents the benchmark estimation results including only the socioeconomic variables. The second column additionally includes the occurrence of floods while the third column includes the occurrence of drought is included. Finally, the last model assesses the effect of both droughts and floods on food security, with all controls.

The estimation results in Table 7 show a positive and significant impact of floods of the previous year on food availability. However, the effect of droughts is not significant on food availability. These effects remain unchanged when both events are included simultaneously. Before interpreting this positive sign of the “flood” variable, it should be noted that the “availability” dimension has more of a quantitative aspect and this can refer to improved harvests or availability. In addition, there is no direct link between food availability and flooding as the first phrase refers to foods that are already available in the local market (regardless of scale), the attic, the family farm, the kitchen and the family store (Ashley, 2016; Reig, 2012) regardless of its origin. If we try to establish a link, the positive sign of floods on the first dimension can be explained by the fact that floods constitute an excess of water, so that if there is a good management, it can increase production. Douglas (2017) promotes the need for good water management for local communities. Households and agricultural populations in the West African Sahel may also anticipate the occurrence of floods. In contrast, droughts have insignificant effects on the first dimension index.

Table 7.

Estimation results for the first dimension of food security: availability.

	Dependent variable: Ind_availability
	(1)	(2)	(3)	(4)
L.flood		5.545** (2.431)		5.538** (2.446)
L.drought			−1.321(5.136)	−1.178(5.002)
log(L.cerealyield)	27.278** (12.143)	26.820** (11.827)	27.263** (12.218)	26.808** (11.901)
log(L.gdcapita)	−11.593(11.093)	−11.329(10.803)	−11.884(11.218)	−11.588(10.926)
log(L.popgrowth)	311.221*** (93.830)	298.690*** (91.534)	316.582*** (96.677)	303.485*** (94.334)
log(L.cerealsimports)	−1.385(8.111)	−0.762(7.903)	−1.621(8.212)	−0.973(8.003)
L.stabpol	−45.719*** (6.706)	−45.620*** (6.530)	−45.820*** (6.759)	−45.710*** (6.582)
L.inflation	1.295** (0.649)	0.852(0.661)	1.258* (0.668)	0.820(0.679)
L.migration	5.293** (2.638)	5.140** (2.570)	5.231* (2.666)	5.085* (2.597)
Country fixed effects	Yes	Yes	Yes	Yes
Observations	90	90	90	90
R ²	0.636	0.660	0.637	0.660
Adjusted R²	0.580	0.601	0.574	0.596
F Statistic	19.238*** (df = 7; 77)	18.402*** (df = 8; 76)	16.637*** (df = 8; 76)	16.161*** (df = 9; 75)
Model	Individual	Individual	Individual	Individual

p < 0.1; **p < 0.05; ***p < 0.01. The notation L. means that the variable has been lagged.

Regarding the control variables, cereal yield has a positive and very significant impact in all specifications. Any increase in cereal yield increases or improves the quantities of cereals available in the countries in all models. In addition, GDP per capita always has an insignificant effect on food availability: over the period, the level of wealth created does not improve food availability. The results also highlight that population growth and migration have significant and positive effects on food availability, in other words, demographic growth caused by waves of internal or external migration and the high birth rate improve food availability in the Sahel countries. Migration can serve as additional agricultural labor for the host population, if land is allocated to them. Thus, it increases food availability with a possibility of crop growth, so about 80% of all this growth goes to the agricultural sector. Similar results are found in Rizk (2019) and Zhu et al. (2018). Moreover, inflation has positive and significant effects without and in the presence of droughts. In the presence of floods, the effect is not significant. In contrast, the coefficients associated to imports are never significant. Finally, stability or absence of violence has negative and very significant effects on food security for all models. This might be explained by the fact that most mass movements take place in large cities and away from fields. Also, political instability does not tend to impact the available quantities of food in the region: the conflicts are often political, and looting remains limited in certain circumstances. Measures can easily be taken in the protection of local markets and food goods.

Our results are globally consistent with the expectations as all control variables have the expected impacts. For the variables of interest, while intuitively, the availability and quality of food for Sahelian populations can be disrupted by the occurrence of droughts and floods, our results can be explained by the fact that households generally have stocks of cereals that cover their consumption of food over a short period (lean period). In addition, some populations benefit from food aid to be able to adjust their food availability.

We now turn to the second dimension, accessibility, which raises an essential question on inequalities, especially in countries where infrastructure is not developed (Ashley, 2016). The estimation results in Table 8 show that floods have negative and very significant effects on food accessibility and this effect remains unchanged even when droughts are also included. Intuitively, floods have detrimental effects on the density of railway lines, prevalence of undernourishment and the number of undernourished people on food accessibility. The lack of road infrastructure also hinders the accessibility of food to populations who may need to travel to nearby markets to obtain food supplies. However, the coefficients associated with droughts are never significant, these events do not impact food accessibility.

Table 8.

Estimation results for the second dimension of food security: accessibility.

	Dependent variable: Ind_accessibility
	(1)	(2)	(3)	(4)
L.flood		−0.223** (0.089)		−0.224** (0.090)
L.drought			0.077(0.192)	0.084(0.185)
log(L.cerealyield)	−0.901** (0.446)	−0.890** (0.431)	−0.907** (0.449)	−0.897** (0.433)
log(L.gdcapita)	2.397*** (0.405)	2.374*** (0.391)	2.411*** (0.409)	2.389*** (0.395)
log(L.popgrowth)	−15.137*** (3.445)	−14.751*** (3.328)	−15.428*** (3.541)	−15.067*** (3.420)
log(L.cerealsimports)	0.198(0.303)	0.203(0.292)	0.215(0.307)	0.222(0.297)
L.stabpol	1.296*** (0.264)	1.320*** (0.254)	1.301*** (0.265)	1.326*** (0.256)
L.inflation	−0.044* (0.024)	−0.026(0.025)	−0.042* (0.025)	−0.024(0.025)
L.migration	−0.001(0.098)	−0.001(0.095)	0.002(0.099)	0.003(0.095)
Country fixed effects	Yes	Yes	Yes	Yes
Observations	84	84	84	84
R ²	0.432	0.478	0.433	0.480
Adjusted R²	0.336	0.381	0.328	0.374
F Statistic	7.709*** (df = 7; 71)	8.024*** (df = 8; 70)	6.685*** (df = 8; 70)	7.074*** (df = 9; 69)
Model	Individual	Individual	Individual	Individual

p < 0.1; **p < 0.05; ***p < 0.01. The notation L. means that the variable has been lagged.

The other variables (cereal yield, GDP per capita, population growth, cereal imports, political stability, migration) have the expected effects on food accessibility. Cereal yield and inflation have negative and mostly significant effects on food accessibility of populations while cereal imports do not have a significant effect on food accessibility of the populations over this period. Therefore, low cereal yields and imports that do not meet domestic demand mean that no population has access to food. The Sahel is characterized by a drop in food production leading to an increase in food prices, thus limiting access to food for a large part of the population. Also, most of the Sahelian population lives below the poverty line on less than a dollar a day (Eboko and Schlimmer, 2020). The permanent rise in prices then makes certain foods sometimes inaccessible and indirectly increases food insecurity of populations who cannot access markets and consume poorly nutritious foods. Prices affect the choice of consumer decisions and the basket of very poor households. Although cereal imports are made to make up for the national food deficit, not everyone has access to them due to low incomes and high food prices in local markets. The effect of GDP per capita is always positive and significant. Thus, the level of wealth favors access to food, even in the presence of weather events such as droughts or floods. Conversely, the coefficients associated with migration are not significant, but population growth has very negative and significant effects on food accessibility. Indeed, a demographic explosion that is not accompanied by an increase in agricultural production or food availability will put pressure on food accessibility.

The results also show that the coefficients associated with political stability and the absence of violence are positive and very significant on food accessibility both in the presence and in the absence of weather events. It therefore plays an important role in explaining food security in its accessibility dimension. Peace and quality of the institutional framework ensure a good climate allowing the populations to eat properly and healthily.

Globally, we then find negative and very significant effects for grain yields, population growth, and food market inflation. As the rate of population growth is increasing faster than the rate of economic growth, population growth and waves of migration can only have adverse effects on food access as this double shock will put pressure on local food availability leading to food shortage because even the host population does not reach food self-sufficiency.

The third dimension, the “use” or “consumption” dimension, shows how people use available foods (Ashley, 2016). This dimension also provides insight into the effect of diet on consumers while making it possible to assess the health status of consumers (Ashley, 2016; FAO, 2014). The consumption of certain food products causes more or less serious health problems for human health. Excessive consumption of a food also causes health problems for the consumer.

The estimation results displayed in Table 9 show that the effect of floods and droughts are negative and very significant on the use or consumption of food. With the floods, many diseases appear due to the lack of infrastructure (drinking water, basic sanitation services) and hygiene, which are the origin of certain diseases, including cholera and malaria, affecting a large part of Sahelian population as also shown by Week and Wizor (2020) in their study on Nigeria and Douglas (2017) in South Asia. Consequently, floods have adverse effects by increasing the percentage of a population without basic drinking water and limiting access to basic sanitation services. In addition, flooding also increases the percentage of children under 5 causing stunting and the percentage of overweight children under 5 and also increasing the prevalence of anemia among women in childbearing age (i.e. women aged 15–49). Conversely, droughts also cause significant problems as households can no longer feed themselves or water their livestock, which have become unsalable, and also the lack of milk. In addition, crops are becoming scarcer, prices are rising and households and individuals are unable to buy the quantity of food necessary for their good health. When the two shocks occur at the same time, the effect of droughts is greater (see Column 4 of Table 9) than that of floods.

Table 9.

Estimation results for the third dimension of food security: use.

	Dependent variable: Ind_utilization
	(1)	(2)	(3)	(4)
L.flood		−0.469* (0.256)		−0.475* (0.252)
L.drought			−0.970* (0.524)	−0.983* (0.515)
log(L.cerealyield)	4.507*** (1.266)	4.546*** (1.247)	4.497*** (1.246)	4.536*** (1.226)
log(L.gdcapita)	1.407(1.156)	1.385(1.139)	1.194(1.144)	1.168(1.126)
log(L.popgrowth)	22.133** (9.780)	23.194** (9.652)	26.071*** (9.861)	27.195*** (9.718)
log(L.cerealsimports)	1.082(0.845)	1.029(0.833)	0.909(0.838)	0.853(0.824)
L.stabpol	−0.243(0.699)	−0.252(0.689)	−0.318(0.689)	−0.327(0.678)
L.inflation	−0.093(0.068)	−0.056(0.070)	−0.120* (0.068)	−0.083(0.070)
L.migration	−0.986*** (0.275)	−0.973*** (0.271)	−1.031*** (0.272)	−1.019*** (0.268)
Country fixed effects	Yes	Yes	Yes	Yes
Observations	90	90	90	90
R ²	0.637	0.652	0.652	0.668
Adjusted R²	0.580	0.592	0.593	0.606
F Statistic	19.273*** (df = 7; 77)	17.798*** (df = 8; 76)	17.826*** (df = 8; 76)	16.774*** (df = 9; 75)
Model	Individual	Individual	Individual	Individual

p < 0.1; **p < 0.05; ***p < 0.01. The notation L. means that the variable has been lagged.

For the control variables, the effect on use is positive and very significant for cereal yield and population growth in all models. Conversely, GDP per capita and cereal imports have insignificant effects since utilization refers to the qualitative aspect of food. For people to make good use of food, it must be available and accessible in the right way. Consumer food price inflation (except in Column 3 of Table 9) and political stability do not have significant effects on food use or consumption either. Finally, migration has negative and very significant effects on the use and consumption of food.

Finally, the fourth dimension refers to the stability and sustainability of the food source over time (Ashley, 2016; Aurino, 2014; FAO, 2014). In other words, it is related to availability, access and predictability. The estimation results displayed in Table 10 show that for the last dimension, the effects of floods and droughts are never significant, including when the two meteorological events occur simultaneously. These results can be explained in the sense that this dimension is related to the temporal aspect of food security, in other words, the horizon, as shown by Oskorouchi and Sousa-Poza (2021) on Afghanistan. It refers to the stability of the first three dimensions.

Table 10.

Estimation results for the fourth dimension of food security: stability.

	Dependent variable: Ind_stability
	(1)	(2)	(3)	(4)
L.flood		−0.519(0.896)		−0.535(0.890)
L.drought			−2.617(1.812)	−2.631(1.820)
log(L.cerealyield)	4.286(4.341)	4.329(4.360)	4.258(4.311)	4.302(4.329)
log(L.gdcapita)	3.964(3.965)	3.939(3.983)	3.388(3.958)	3.359(3.975)
log(L.popgrowth)	−51.300(33.541)	−50.128(33.748)	−40.683(34.109)	−39.417(34.318)
log(L.cerealsimports)	−7.580** (2.899)	−7.638** (2.914)	−8.046*** (2.897)	−8.109*** (2.911)
L.stabpol	2.905(2.397)	2.896(2.408)	2.705(2.385)	2.694(2.395)
L.inflation	−0.128(0.232)	−0.087(0.244)	−0.201(0.236)	−0.159(0.247)
L.migration	−1.980** (0.943)	−1.966** (0.948)	−2.104** (0.940)	−2.090** (0.945)
Country fixed effects	Yes	Yes	Yes	Yes
Observations	90	90	90	90
R ²	0.199	0.202	0.220	0.224
Adjusted R²	0.074	0.066	0.087	0.079
F Statistic	2.726** (df = 7; 77)	2.406** (df = 8; 76)	2.679** (df = 8; 76)	2.402** (df = 9; 75)
Model	Individual	Individual	Individual	Individual

p < 0.1; **p < 0.05; ***p < 0.01. The notation L. means that the variable has been lagged.

Cereal yield, GDP per capita and political stability do not have a significant effect on food stability. In the Sahel, low cereal yields do not ensure stability and do not allow the population to have long-term access to cereals. The reasoning is also valid for GDP per capita, with very low daily incomes, populations will not be able to access food at all times because households must limit the number of meals according to their financial capacities. Furthermore, population growth and inflation also have insignificant effects on food stability over time. These results confirm our results on the three previous dimensions. Finally, migration has negative and significant effects on the stability of food availability for Sahelian populations. Intuitively, the arrival of new people affects food availability by increasing instantaneous consumption, thus limiting the build-up of food stocks and subsequent consumption by local and host populations (Zhu et al., 2018). Cereal imports have a negative and significant effect on food stability: food security cannot be built entirely with grain imports. Within the framework of a food security strategy, cereal imports can only be a compensation tool.

Conclusions and discussion

This article has examined the link between weather events and food security in West African Sahel countries, controlling for other determinants. A panel data model made it possible to isolate the effect of climatic shocks (floods and drought) on the four dimensions of food security (availability, accessibility, stability and use) for the period from 2001 to 2017. The results show that, overall, climate shocks have an impact on food security in countries. Specifically, food insecurity in these countries over this period is accentuated by the floods, through its impact on accessibility and use. It also shows that the level of wealth creation and inflation are key determinants of food security. Political stability and the absence of violence are more than an imperative in ensuring a food security situation. Finally, we show that migration puts pressure on local food demand and does not improve food security.

Overall, our results point a multitude of recommendations for all actors (political decision-makers, international and national institutions) to increase and improve the level of food security toward a sustainable agrifood system in the Sahel countries.

First, our results point out that food availability is not impacted negatively by climatic events but is not a sufficient condition for food security, yet necessary. The first dimension must be satisfied because the gap between demand and local food supply is a source of food insecurity and also of social conflict. In this context, any improvement in the agricultural sector will have direct and positive impacts on the food security of populations. The development of the irrigated sector could be a strategy for reducing poverty, adapting to the climate and promoting food security (Kafle and Balasubramanya, 2022). The role of irrigation in poverty reduction is proven because farmers especially smallholders will not only focus on the 3-month rainy season (Yobom, 2020) but will be economically active throughout the entire period of the year (Partey et al., 2018). Access to irrigation water via the motor pump has increased household savings and informal social insurance in the form of transfers in northern Mali (Dillon, 2008; Jambo et al., 2021). In addition, year-round vegetable production facilitated by canal irrigation in northern Senegal increased vitamin A and C intake and decreased the incidence of wasting in adults and older children (Bénéfice and Simondon, 1993; Akpaki et al., 2020).

Second, inflation, with its overall negative impact of food security must be brought under control, so that income and the standard of living of the population improve. The ease of food accessibility could be boosted by the reinforcement of sustainable social safety nets. States can set up compensation funds to subsidize basic necessities as most households devote a large part of their income to this type of daily purchase.

Third, our results clearly show that political stability and the absence of violence improve food security. As a result, governments must ensure the security of populations and strengthen stability within their territory and at their borders. Without political stability, people are not able to develop income-generating activities and practice agriculture to ensure decent food. The way conflict-related disruptions in global food and fertilizer markets affect prices and availability is key to understanding the overall impact on global food security (Ben Hassen and El Bilali, 2022). This relation has been proven important elsewhere. For example, Soffiantini (2020) showed an existential link between food insecurity and political instability during the Arab Spring, particularly in Egypt, Syria and Morocco.

Fourth, migration has a positive impact on food availability but a negative impact of the third and fourth dimensions. Establishing adequate social arrangements can mitigate the effects of migratory flows caused by conflict and climate change on food availability. The swelling of the local population by the arrival of waves of migrants only amplifies the pressure on local food demand. Refugee reception and integration agencies must, in common agreement with local governments, internalize refugee flows. Note that our results on the positive effect of migration on food availability are widely shared in the literature. Managing refugee flows through land grants has positive economic effects, which has also been shown by Zhu et al. (2024). Another dimension on the impact of demography is population growth. States must be able over time of transforming population growth into opportunities. These generations must be a support and not a burden for the sustainability of economic growth in the long term. Well-nourished children and youth are productive and will support economic activities (McGovern et al., 2017; Rizk, 2019).

As a general conclusion, we point out in a region where 80% of the population lives off agriculture and where there are no social safety nets, the development of the agricultural and food system is a political and economic obligation. It is about renewing and building a new agricultural system. To feed all the Sahelian demands, a diversity of structures and agricultural systems is necessary. Investment in agronomic research by developing cereals that can be grown in Sahelian climatic conditions to increase national agricultural production. Finally, the establishment of information on the agricultural calendar and the training of farmers will make it possible to avoid certain agricultural losses while improving harvests and therefore food security. Actors operating in the Sahel must act on the four dimensions strategically in their fight against food insecurity to achieve sustainable development objectives. All public policy actions must be applied simultaneously on the four dimensions to achieve all food security objectives and move toward the Millennium Development Goals.

Supplemental Material

sj-docx-1-jas-10.1177_00219096231225949 – Supplemental material for Food Security and Weather Events: A Multidimensional Analysis in the West African Sahel for 2001–2017

Supplemental material, sj-docx-1-jas-10.1177_00219096231225949 for Food Security and Weather Events: A Multidimensional Analysis in the West African Sahel for 2001–2017 by Oudah Yobom and Julie Le Gallo in Journal of Asian and African Studies

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Julie Le Gallo

Supplemental material

Supplemental material for this article is available online.

Notes

Author biographies

Oudah Yobom is an applied and development economist whose research focuses on the economics of climate change and the economics of agricultural sectors in the Sahel. Issues related to poverty and the resilience of agricultural households are essential aspects of his work.

Julie Le Gallo is an economics professor at Institut Agro Dijon (France) and an applied economist whose research focus on climate migrations, residential mobility, local public economics, land and housing prices and agricultural and environmental economics.

References

Ahmad Munir

Iqbal

(2016) Impact of farm households’ adaptations to climate change on food security: Evidence from different agro-ecologies of Pakistan. The Pakistant Development Review 55(4): 561–588.

Akpaki

Galibois

Sall

, et al. (2020) Assessing the food availability and food insecurity situation among communities of Matam region, Senegal. Ecology of Food and Nutrition 59(4): 367–386.

Asfaw

(2015) Adaptation to Climate Risk and Food Security: Evidence from Smallholder Farmers in Ethiopia. Rome: FAO.

Ashley

(2016) Food Security in the Developing World. Cambridge, MA: Academic Press.

Aurino

(2014) Selecting a Core Set of Indicators for Monitoring Global Food Security. A Methodological Proposal (FAO Food and Nutrition Series). Rome: FAO.

Babu

Gajanan

Sanyal

(2014) Food Security, Poverty and Nutrition Policy Analysis: Statistical Methods and Applications. Cambridge, MA: Academic Press.

Belloumi

(2014, April) Investigating the linkage between climate variables and food security in ESA countries. AGRODEP working paper no. 0004. Washington, DC: International Food Policy Research Institute.

Ben Hassen

El Bilali

(2022) Impacts of the Russia-Ukraine war on global food security: towards more sustainable and resilient food systems? Foods 11(15): 2301.

Bénéfice

Simondon

(1993) Agricultural development and nutrition among rural populations: a case study of the middle valley in Senegal. Ecology of Food and Nutrition 31(1–2): 45–66.

10.

Bilan

Lyeonov

Stoyanets

, et al. (2018) The impact of environmental determinants of sustainable agriculture on country food security. International Journal of Environmental Technology and Management 21: 289–305.

11.

Blanc

Schlenker

(2017) The use of panel models in assessments of climate impacts on agriculture. Review of Environmental Economics and Policy 11: 258–279.

12.

CILSS (2019) Sahel et Afrique de l’ouest, situation alimentaire et nutritionnelle. https://www.food-security.net/wp-content/uploads/2019/12/WA-regional-profile_FR.pdf

13.

Davis

Winters

Carletto

, et al. (2010) A cross-country comparison of rural income generating activities. World Development 38(1): 48–63.

14.

Demeke

Keil

Zeller

(2011) Using panel data to estimate the effect of rainfall shocks on smallholders food security and vulnerability in rural Ethiopia. Climatic Change 108(1–2): 185–206.

15.

Dillon

(2008) Access to Irrigation and the Escape from Poverty: Evidence from Northern Mali (Vol. 782). Addis Ababa: International Food Policy Research Institute.

16.

Douglas

(2017) Flooding in African cities, scales of causes, teleconnections, risks, vulnerability and impacts. International Journal of Disaster Risk Reduction 26: 34–42.

17.

Eboko

Schlimmer

(2020) COVID-19: l’Afrique face à une crise mondiale. Politique Étrangère 4: 123–134.

18.

Ervin

Gayoso de Ervin

(2019) Household Vulnerability to Food Insecurity in the Face of Climate Change in Paraguay. Rome: FAO.

19.

FAO (2014) State of Food Insecurity in the World 2013: the Multiple Dimensions of Food Security. Rome: FAO.

20.

Food Security Information Network (FSIN) (2019) Global report on food crises. https://www.fsinplatform.org/report/global-report-food-crisis-2019/

21.

Husson

Lê

Pagès

(2016) Analyse de données avec R. FAO.

22.

Jambo

Alemu

Tasew

(2021) Impact of small-scale irrigation on household food security: evidence from Ethiopia. Agriculture & Food Security 10(1): 1–16.

23.

Jones

Ngure

Pelto

, et al. (2013) What are we assessing when we measure food security? A compendium and review of current metrics. Advances in Nutrition 4(5): 481–505.

24.

Kafle

Balasubramanya

(2022) Reducing food insecurity through equitable investments in irrigation: the case of Niger. Journal of the Agricultural and Applied Economics Association 1: 494–515.

25.

Karfakis

Knowles

Smulders

, et al. (2011, September) Effects of global warming on vulnerability to food insecurity in rural Nicaragua. ESA working papers no. 289003. Food and Agriculture Organization of the United Nations, Agricultural Development Economics Division (ESA).

26.

Kassambara

(2017) Practical Guide to Principal Component Methods in R: PCA, M (CA), FAMD, MFA, HCPC, Factoextra (Vol. 2). Sydney, NSW, Australia: STHDA.

27.

Kinda

Kere

(2016) Climate Change and Food Security: Do Spatial Spillovers Matter? New York: Mimeo.

28.

McGovern

Krishna

Aguayo

, et al. (2017) A review of the evidence linking child stunting to economic outcomes. International Journal of Epidemiology 46(4): 1171–1191.

29.

Masipa

(2017) The impact of climate change on food security in south Africa: current realities and challenges ahead. Jàmbá Journal of Disaster Risk Studies 9: 411.

30.

Maxwell

(1996) Measuring food insecurity: the frequency and severity of “coping strategies..” Food Policy 21(3): 291–303.

31.

Nyaga

Doppler

(2009) Combining principal component analysis and logistic regression models to assess household level food security among smallholder cash crop producers in Kenya. Quarterly Journal of International Agriculture 48(1): 5–23.

32.

Oskorouchi

Sousa-Poza

(2021) Floods, food security, and coping strategies: evidence from Afghanistan. Agricultural Economics 52(1): 123–140.

33.

Partey

Zougmoré

Ouédraogo

, et al. (2018) Developing climate-smart agriculture to face climate variability in West Africa: challenges and lessons learnt. Journal of Cleaner Production 187: 285–295.

34.

Reig

(2012, July) Food security in African and Arab countries: a review of the topic and some suggestions for building composite indicators with principal components analysis. Technical report Working Papers 1210. Department of Applied Economics II, Universidad de Valencia.

35.

Reyes

Bancolita

Lor Leyso

, et al. (2014) Impacts of Climate Change on Household Food Security in the Philippines. Rome: FAO.

36.

Rizk

(2019) Does demographic transition matter for economic growth? Evidence from Egypt. The Journal of North African Studies 24(6): 1012–1035.

37.

Soffiantini

(2020) Food insecurity and political instability during the Arab Spring. Global Food Security 26: 100400.

38.

Week

Wizor

(2020) Effects of flood on food security, livelihood and socio-economic characteristics in the flood-prone areas of the core Niger Delta, Nigeria. Asian Journal of Geographical Research 3(1): 1–17.

39.

Yobom

(2020) Climate change and variability: empirical evidence for countries and agroecological zones of the Sahel. Climatic Change 159(3): 365–384.

40.

Yobom

Le Gallo

(2022) Climate and agriculture: empirical evidence for countries and agroecological zones of the Sahel. Applied Economics 54(8): 918–936.

41.

Zhu

Gupta

Filipski

, et al. (2024) Economic impact of giving land to refugees. American Journal of Agricultural Economics 106: 226–251.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.27 MB