AGRICULTURAL ECONOMICS Carbon emission and global food ...

PENCIL Publication of Agricultural Sciences Vol. 2(1):7-24 ISSN: 2408-5545 Available a t: www.pencila cademi cpress.org/ppas (c)2015 PENCIL Academi c Press

AGRICULTURAL ECONOMICS

Research

Carbon emission and global food security: A cross country analysis Ajay Kumar 1*, Mokbul Morshed Ahmad2 and Pritee Sharma3 Authors' Affiliations

ABS TRACT

1 Institute

This study was carried out to investigate the influence of carb on emissions and other socio-economic factors on global food security. Also, it creates global food security index (GFS I) for thirty one cross country with different income groups from various regions of the World. In this study, Composite Z-in dex technique was employed in order to generate GFS I using several key determinants of food security. Subsequently, multiple linear regression models were also employed in order to assess the impacts of carb on dioxide emission and socio-economic factors on the constructed GFSI using country wise panel data. The descriptive results of the study show that high income countries are the most food secured due to high cereal productivity, per capita land under cereal crops, per capita arable land, and high per capita gross domestic product. Developed countries have negligible poverty, constant population growth rate and le ss dependency on cereal import. Low and lower income group countries are in food insecurity trap due to rapid p opulation growth, high infant mortality, high volatility in per capita food production and incidence of food-deficit. This study emphasized that lower middle and low income countries need to increase cereal production, in order to sustain food security. Based on the findings of this study, it is suggested that if the world's population grows at the current rate, then food insecurity would be more alarming in the near future. Empirical results recognized that per capita CO2 emission has negative impact on global food security negatively. Therefore, this study strongly recommends the necessity of the world's econom ies to develop alternative scientific techniques to abate greenhouse gas.

of Rural Management, Anand, Gujarat-388001, India. 2 Regional

and Rural Development Planning, School of Environment, Resources and Development, Asian Institute of Technology, Pathumthani-12120, Thailand. 3 School

of Humanities and Social Sciences, Indian Institute of Technology Indore, Madhya Pradesh – 452017, India.

*Corresponding author E-mail: [email protected], [email protected]. Tel: +91-7566569201. Accepted: 8 th November, 2015. Published: 1st December, 2015.

Key words: World, cross country, carbon emission, global food security, GFSI, poverty.

INTRODUCTION Climate variability has brought a complex, multidimensional and several challenges for humanities in term of several socio-economic variables at global level (Ranganathan, 2010; Arndt et al., 2012). The most effect of climate change would be increase in temperature due to the rising greenhouse gases (GHG), such as carbon dioxide, methane, ozone, nitrous oxide,

and chlorofluoro carbons (Mall et al., 2006; Gadgil, 1995). Scientific research community’s studies have claimed that the patterns of climatic factors are changing due to the rising quantity of GHG emissions, which may put many social, biological and geophysical systems at danger (Gadgil, 1995; Zhang, 2008; Mall et al., 2006; Girardet and Bree, 2009). The increasing temperature,

8. PENCIL Pub. Agric. Sci.

fluctuation in rainfall pattern, floods, hail storm, drought, rising sea level, melting of glaciers, heavy wind, cyclone, earthquake and deforestation are clear signs of climate change (Gadgil, 1995). GHG emission is increasing in the atmosphere as carbon dioxide concentration was demonstrated to reach 280 ppm at the end of the preindustrial period (1850), and after which its concentrations in the atmosphere have been shown to lead at the rate of 1.5 to 1.8 ppm per year (Mall et al., 2006). Pant (2009) also observed that since 1970, the amount of carbon dioxide, methane and nitrous oxide has increased by approximately 31, 151 and 17%, respectively in the atmosphere. Rapid industrialization, high economic growth and development, high population growth rate, fast urbanization, building and road construction, degradation of forest area, mechanization, application of fertilizer and pesticides in agriculture, deforestation and waste production are putting enormous pressure on scare natural resources. These anthropogenic activities have increased the qualitative and quantitative degradation of land, water, air, biodiversity, bio-resources and ecosystem services (Aggarwal, 2008; Girardet and Bree, 2009). In addition, climate change has increased the difficulties and challenges of the developing and low income countries (Arndt et al., 2012). Most developing countries are located at lower latitudes and as such, their economies are faced with the problem of reduction of crop yields due to climate change (Lee, 2009; Girardet and Bree, 2009; Nath and Behera, 2011). On the contrary, developed countries are located at higher latitudes, and as such, many of them will get the benefits of high crop yields. Moreover, developing countries would be unable to cope with climate change due to low level of technology, low financial capacity of farmers to mitigate the adverse effect of climate change, high dependency of the population on agriculture and high dependency of farming on rainfall (Mall et al., 2006). More than 50% of farmers in developing countries are subsistence and marginal, they mainly produce for their own consumption (Nath and Behera, 2011). To sustain agricultural productivity and food security in the era of climate change is an emerging issue particularly for large agrarian economies(Julia, 2005; Clark et al., 2010; Greg, 2011). There are about 1 billion populations who are food insecure in the world (Burke and Lobell, 2010). Approximately, more than one sixth the populations (around 900 million) are in the stage of chronic hunger (Johnson, 2009), and it has been estimated that a child dies somewhere in every six second due to unavailability of food (Hollaender, 2010). Further, climate change imposed serious threats to the

Kumar et al. (2015)

agricultural production system and food security to the world's most vulnerable people (Clark et al., 2010). Climate change may affect the livelihood of 200 million people and their families who are engaged in fishing and aquaculture activities in the world (Greg et al., 2011). Also, it could be one of the critical challenges in the achievement of sustainable development in most of developing economies(Clark et al., 2010). In this regards, numerous studies have shown that increase in temperature has negative association with crop yields, livestock, and fishery; and high temperature increases susceptibility to the new diseases (Ranuzzi and Srivastava, 2012). In developing countries, agricultural output and yields may decline by 20 and 15%, respectively (Masters, 2010; Lee, 2009), whereas output in industrial countries by 2080 is expected to decrease by 6% due to climate change(Greg et al., 2011). Horowitz (2009)'s estimation also point out that increase in 1 0C temperature may result in 3.8% decline in world's GDP. Hence, climate change would be a critical challenge to agricultural productivity and food security (locality of production, supply, volume, and quality) by 2080 (Masters, 2010). Drawback of earlier studies and objective of the study Existing studies have shown that climate change has a negative and significant implication on agricultural productivity and food security across economies. Gbetibouo and Hassan (2005), Zhai and Zhuang (2009), Zhai et al. (2009), Horowitz (2009), Lee (2009), Masters et al. (2010); Greg (2011) and Alam (2013) assessed the impact of climate change on gross domestic product (GDP) in various economies. Most studies have examined the economic impact of climate change on agriculture productivity, whereas, few studies included food-grain productivity as proxy for food security in different regions of the world (Mendelsohn et al., 1994; Deressa et al., 2005; Seo and Mendelsohn, 2007; Kurukulasuriya and Mendelsohn, 2008; Seo and Mendelsohn, 2008; Yu et al., 2010; Ajetomobi et al., 2011; Mendelsohn et al., 2011; Gupta et al., 2012). These studies concluded that climate change has huge negative impact on food security in agrarian economies. These study also forecasted that food security also would deteriorates in era of climate change. However, analysis based on two or three food-grain crops may be unrealistic to provide strong evidence as to whether food security would increase or decrease in the presence of climate change. Moreover, food security of a region is


Kumar et al. (2015)

not associated with food-grain crop only. The availability of commercial crops also have significant impacts on food security as production of commercial crops create employment opportunities and economic capacity of working labors in agricultural. In addition, there are many complementary factors that are significantly related with food security, that is, income of the people, per capita food-grain availability, employment opportunities, high population growth, urbanization and poverty etc. Thus, only few studies have considered these factors in the estimation of food security at global level. To account for this drawback, the present study tends to assess the association of global food security with carbon dioxide emission as proxy for climate change (Alam, 2013)and other socio-economic variables in the selection of set of world's economies. In this study, the following objectives were achieved: (1) to generate global food security index (GFSI) using its key components (that is, food availability, stability and accessibility) for cross country during 1990-2009. This investigation would be useful to identify which country is more food insecure than other. (2) To assess the association of GFSI with its components. This estimation could be crucial to determine which components of food security play an imperative role in the improvement of food security. (3) To empirically investigate the impact of per capita carbon emissions and socio-economic factors on constructed GFSI. This exercise will facilities the consistent and reasonable policies that will help reduce food insecurity at global level. (4) To examine the relationship between GFSI and poverty.

considered as sole determinant of food security (Dev and Sharma, 2010; Yu and You, 2013). Another importance of increasing agricultural productivity would be beneficial to poverty eradication and reduction of chronic hunger, as well (Salami, 2011; Prabha et al., 2010). Therefore, agriculture is a crucial sector that aids the achievement of reliable access to adequate, affordable, nutritional and sufficient food to avoid chronic hunger. It also builds foundation for national prosperity, social welfare and human well-being. Food secure individual is likely to have high productivity and contributes more to economic growth and development (ADB, 2012). Furthermore, food security is a crucial factor in the achievement long-term sustainable growth and development of a country (ADB, 2012). The development of a country depends upon the health situation of its population that lives in its national boundary, whereas the health of the population depends on sufficient food supply. Hence, proper health conditions of a population is useful to improve the marginal productivity of workers. Thus, food losses due to any reason may have negative implications on food security, food quality and safety, and economic growth and development (Ramasamy and Moorthy, 2012). There are strong relationship between agricultural productivity, food security, hunger and poverty, which reflect the malnourishment level of people (Hollaender, 2010). Subsequently, climate change, agricultural productivity, food security, hunger, undernourishment and poverty are associated with each other (Hollaender, 2010; Kramer, 2007).

AGRICULTURAL SECURITY

Food security and its components

PRODUCTIVITY

AND

FOOD

Food security of a country/region directly depend upon agricultural productivity (Mall et al., 2006; Masters et al., 2010; Gregory et al., 2012). Agricultural is sole sector that plays a crucial role in the provision of adequate food to humanities (Dev and Sharma, 2010). However, food security may not be defined by agriculture sector only. As food security is a multidimensional, multi-processing and complex phenomenon in an economy (Schmidhuber and Tubiello, 2007; Burke and Lobell, 2010). It directly and indirectly depends upon various factors such as age, sex, occupation, increasing demand of food-grain product, vegetarian and non-vegetarian diet, low productivity of land, traditional technology of cultivation and low education level of farmers, degradation of arable area, decline in ground water and climate change (Shakeel et al., 2012). Further, agriculture sectors are

Food is something that gives us the energy to function and to keeps us alive. Food security is a specific situation in which 'all individual in a society or community, at all times, has physical, social and economic access to sufficient, safe and nutritious food that meets their dietary needs and food preferences for an active and healthy life' (FAO, 2002). Food security comprises of four components (that is, food availability, stability, accessibility and utilization) (ADB, 2012). Food availability refers to the existence of food stocks for consumption through domestic production and commercial imports of food or food aid (Salami, 2011; ADB, 2012; Greg et al., 2011). This component of food focuses on the production of food, livestock and fish in a particular region (FAO, 2009; Burke and Lobell, 2010; Salami et al., 2011). Food accessibility specified the household's ability to acquire food through the


production or purchase of adequate quality and quantity of food to meet the people's nutritional requirements (Greg et al., 2011; Burke and Lobell, 2010). This component is allied with economic capacity of the people to acquire food (FAO, 2009). It is determined by physical and financial resources of households. It also depends on the food prices, market expansion, social and political factor (Greg et al., 2011; Salami et al., 2011). Food stability indicates the continuity in food availability and accessibility in domestic market (FAO, 2008). It is associated with stability in supply and demand of food (FAO, 2009). Food utilization is defined in terms of actual nutritional condition of food. It implies that people must be in position to absorb and metabolize proper nutrients from consumed food. Food utilization is associated with health and actual nutritional factors as per scientific research (FAO, 2008; FAO, 2009; Salami, 2011). In summary, food availability is related to agricultural production, distribution and exchange. However, food accessibility is associated with affordability, allocation and preference. Food utilization is linked with nutritional and social value, and food safety (Gregory et al., 2012). Food stability reveals the equilibrium situation of the demand and supply of food in domestic market. Climate change and components of food security Climate change may affect food security through variation in agricultural production (Schmidhuber and Tubiello, 2007; Girardet and Bree, 2009; Joshi, 2012; Joshi, 2012). Subsequently, economic growth, income distribution and agricultural demand would be adversely affected during variation in climatic factors (Schmidhuber and Tubiello, 2007; Greg et al., 2011). Food security may adversely cause decline in food-grain productivity due to change in rainfall pattern, floods, warmer or cooler temperature (Gregory et al., 2012). Climate change affects all components of food security (Greg et al., 2011; Ranuzzi and Srivastava, 2012; FAO, 2008). Since climate change has negative impact on crop yields, crop pests, soil fertility and water-holding properties (Greg et al., 2011; Joshi, 2012), therefore, food availability may decline. Physical, economic and social accessibility of food would be adversely affected due to climate change (Greg, 2011). Thus food accessibility power of people may decrease due to decline or damage in agricultural production. This means that large number of individuals may be unable to acquire food according to their need. High fluctuation in temperature, rainfall pattern and humidity may result in

Kumar et al. (2015)

decline of actual nutritional contents and quality of food. Government food price policies also would be demoralized in domestic market due to climate change. Climate change has negative impacts on food supply and increase food prices due to scarcity of food in local market (Joshi, 2012; Pandey, 2009). Definition of food security index There are more than 450 parameters and 200 definitions that describes the state of food security in existing literature (Hoddinott, 1999). However, there is no exact, specific or scientific definition of food security index(FSI) in scientific literature. In the present study, global food security index (GFSI) is defined as a composite index that includes three components of food security (that is, food availability, stability,and accessibility). The estimated index indicates the rank across country, which creates the consciousness to policy makers and economic agents to take precautionary action to ensure food security. It also provides the cross comparison between more than two countries/regions/individuals. It may be estimated at macro level (that is, country-wise, region-wise, and state-wise) and micro level (that is, individuals and households). In the case of micro level, it may be created at household level in a particular region and indicates the food inequality in community or society. Even though, there is one major criticism about this index, it puts arbitrary weights and ranking, which always change with every minor data revision. It is useless for inter-temporal comparisons. Regardless of these, it is a very crucial parameter to obtain the insightful idea towards food security at macro and micro level (Ye et al., 2013; Sajjad et al., 2014). MATERIALS AND METHODS Study area In order to generate global food security index (GFSI) with its components, such as availability, stability and accessibility of food, thirty one cross county with different income groups of the world were included in the present study. In this study, based on World Bank's (2011) definition, two-high income, eleven-upper middle income, and eighteen-lower middle income countries were included in this investigation. The details of the selected countries were undertaken as: High Income Countries:- Hungary and Poland; Upper Middle Income


Countries:- Brazil, China, Colombia, Costa Rica, Ecuador, Malaysia, Mexico, Panama, Romania, Thailand and Turkey; Lower Middle Income Countries:- Bolivia, Egypt, El Salvador, Guatemala, Honduras, India, Indonesia, Moldova, Morocco, Pakistan, Paraguay, Philippines, Sri Lanka, Ukraine, Nigeria, Kyrgyz Republic, Bangladesh and Tanzania. Source of data The present study, included 30 years data as countrywise panel during 1990-2009. All the required data for a given time period were derived from World Bank, Food and Agriculture Organization of the United Nations (FAO), World Meteorological Organization (WMO), United Nations Framework Convention on Climate Change (UNFCCC), Intergovernmental Panel on Climate Change (IPCC), Yale Center for Environmental Law and Policy (YCELP) (Yale University), Center for International Earth Science Information Network (CIESEN) (Colombia University), and International Food Policy Research Institute (IFPRI). Linear interpolation and graphical projection methods were used in order to incorporate the missing values in the undertaken time series data (Mondal et al., 2014). Generation of global food security index (GFSI) There are many methods used in the estimation of food security index (FSI) in literatures. Shakeel et al. (2012) and Rukhsana (2011) estimated the district-wise FSI using Composite Z-index technique in Uttar Pradesh (India). Kumar and Sharma (2013) generated state-wise FSI for empirical analysis in India. Shakeel et al. (2012) and Rukhsana (2011) employed the simple descriptive method to generate FSI in Uttar Pradesh (India). These studies included most of variables of food security (that is, food availability, stability, and accessibility).Omotesho et al. (2006) and Ibrahim et al. (2009) estimated the FSI using the actual content on nutrients method in Nigeria. In Ethiopia, FSI is generated using the Principle Component Analysis, which included human, social, physical, financial and natural capital (Demeke et al., 2011). In the present study, simple descriptive technique was used in order to generate the Global Food Security Index (GFSI)(EIU, 2012, 2013, 2014, 2015 1; Kumar and Sharma, 2013). While, three components of food security, such as food availability, accessibility and stability, were included in the current study. Food utilization component was not included in this study due to unavailability of information regarding food

Kumar et al. (2015)

utilization. The technique used in this study was adopted from The Economist Intelligence Unit(EIU Report 2012, 2013, 2014, 2015 1). If the variables favour the global food security than Composite Z-index(CI), the estimate is given as: CIi = {[x i – Min(x)]/[Max(x) – Min(x)]}

(1)

Where, CI is the Composite Z-index; Min(x) and Max(x) are the lowest and highest values in each series of data in across country for a specific variable, respectively; i indicates the ith variable. If the variables are negatively associated with global food security than CI, the following estimate is given: CIi = {[x i – Max (x)]/[Max (x) – Min (x)]}

(2)

In Equation (2) all values are transformed into a positive number on a scale of 0-1, in order to make direct comparison with each other (EIU, 2013, 2014, 2015).1 The final GFSI for a particular country is estimated as: GFSIc = ∑CI i/n

(3)

Where, GFSI is the global food security index; and GFSI value lies between 0 to 1; 1 (one) indicates the most food secure country and 0 (zero) represents the highest food insecure country; c is the specific country and n is the total number of variables (EIU, 2013, 2014, 2015)1. Thirty one cross country with different income groups were included to generate GFSI during 1990-2009 (See Appendix: A). The following variables were used in order to estimate the GFSI: (i) Food availability (AVAF): (1) Arable land (in % of total land area) (ARL), (2) Per capita electric power consumption (in kWh) (PCEPC), (3) Per capita energy use (inkg of oil equivalent) (PCEU), (4) Per capita food production variability (PCFPV) (in %), (5) Per capita dietary energy supply (kcal/caput/day) (in calorie)(PCDES), and (6) Per capita depth of food-deficit (kcal/caput/day) (in calorie)(PCDFD). (ii) Food stability (STAF): (1) Cereal yield (in kg per hectare) (CY), (2) Forest area (in% of total area) (FA), (3) Employment in agriculture (in % of total employment) (EAGS), (4) Per capita land under cereal production (in hectare)(PCLUCP), (5) Percentage of arable land equipped for irrigation (IRL)(in %), and (6) Per capita carbon dioxide (CO2) gas emissions (PCCDGE)(in metric tons). (iii) Food accessibility (ACCF): (1) Per capita arable


land (in hectare) (PCAL), (2) Per capita GDP (in current US$) (PCGDP), 3) Female population (in % of total population) (FP), (4) Ratio of urban population with rural population (in ratio)(UR), (5) Infant mortality rate (in per 1000 live births) (IMR), (6) Labor participation rate(in % of total population aged 15+)(LPR), 7) Poverty gap (in % at $2 a day) (PPP) (PG), and (8) Population growth rate (annual %) (PGR). FORMULATION OF EMPIRICAL MODELS Few studies have empirically investigated the association of socioeconomic variables and climatic factors with estimated food security index (FSI). Demeke et al. (2011) assessed the impact of climatic and socio-economic factors on constructed FSI in Ethiopia. In the present study, multiple linear regression models was applied in order to assess the impacts of carbon dioxide emission and socio-economic factors on constructed GFSI using country-wise panel data. The proposed model assumes that GFSI is a function of food availability, stability and accessibility (Kumar and Sharma, 2013). This can be written as: GFSI = f {(Food Availability), (Food Accessibility), and (Food Stability)} Or GFSI = f {(Food Availability: food grain production, calorie availability, etc.), (Food Accessibility: percentage of main worker, poverty, income of people, agricultural credit etc.), (Food Stability: food grain yield, consumption of fertilizers, irrigated area, etc.)} (4) Hence, Equation (4) takes the following empirical form: (GFSI)ct = α0 + α1 (AVAFI)ct +α2 (STAFI)ct +α3 (ACCFI)ct +µi (5) Where, GFSI is the estimated global food security index for cross country; c represent the cross country and t represent the time period (1990-2009). AVAFI, STAFI and ACCFI are the estimated indices for food availability, food stability and food accessibility, respectively; α0 is the constant term; α1, α2, and α3 are the regression coefficient for the corresponding variables; and µi is the error term. By incorporating all variables of each component into Equation (5), it can be specified as: (GFSI)ct = f{(ARL)ct, (PCEPC)ct, (PCEU)ct, (PCFPV)ct, (PCDES)ct, (PCDFD)ct, (CY)ct, (FA)ct, (EAGS)ct, (PCLUCP)ct, (IRL) ct,(PCCDGE)ct, (PCAL)ct, (PCGDP)ct, (FP)ct, (UR)ct, (IMR)c t, (LPR)c t, (PG)ct, (PGR)ct} (6)

Kumar et al. (2015)

Where, ARL, PCEPC, PCEU, PCFPV, PCDES, PCDFD, CY, FA, EAGS, PCLUCP, IRL, PCCDGE, PCAL, PCGDP, FP, UR, IMR, LPR, PG, and PGR are arable land (% of total land area), per capita electric power consumption (in kWh), per capita energy use (in kg of oil equivalent), per capita food production variability(in %), per capita dietary energy supply (kcal/caput/day)(in calorie), per capita depth of food-deficit (kcal/caput/day)(in calorie), cereal yield (in kg per hectare), forest area (in % of total cultivable area), employment in agriculture (in % of total employment), per capita land under cereal production (in hectare), percentage of arable land equipped for irrigation(in %), per capita carbon dioxide (CO 2) gas emissions (in metric tons), per capita arable land (in hectare), per capita gross domestic product (GDP)(in US $), female population (in % of total population), ratio of urban population to rural population(in ratio), infant mortality rate (per 1000 live births)(in number), labor participation rate(in %), poverty gap(in %), and population growth rate (in %), respectively as shown in Equation (6). By applying multiple linear regressions model, Equation (6) is specified as: (GFSI)ct = β0 + β1(ARL)ct+ β2(PCEPC)ct+ β3(PCEU)ct+ β4(PCFPV)ct+ β5 (PCDES)ct+ β6(PCDFD)ct+ β7(CY)ct+ β8(FA)c t+ β9(EAGS)ct + β10(PCLUCP)ct+ β11 (IRL)ct + β12 (PCCDGE)ct+ β13(PCAL)ct + β14(PCGDP)ct+ β15(FP)c t+ β16(UR)ct+ β17(IMR)ct + β18(LPR)ct+ β19(PG)ct + β20(PGR)ct +µi (7) Where, β0 is the constant coefficient; β1 to β20 are the regression coefficient for the corresponding variables and µi is the error term (Demeke et al., 2011; Kumar and Sharma, 2013). Other variables are defined in Equation (6). To assess the impacts of socio-economic factor on each components of GFSI, the following regression models were incorporated: (AVAFI)ct = Δ0 + Δ1 (ARL)ct + Δ2 (PCEPC)ct + Δ3 (PCEU)ct + Δ4 (PCFPV)ct + Δ5 (PCDES)ct + Δ6 (PCDFD)c t+ ui (8) Where, AVAFI is the food availability index, Δ0 is the constant term; Δ1 to Δ6 are the regression coefficient for corresponding variables; and ui is the error term in Equation (8)(Kumar and Sharma, 2013). Descriptions of other reaming variables are stated in Equation (6). (STAFI)c t = λ 0+ λ 1(CY)ct + λ 2(FA)ct + λ 3(EAGS)ct + λ 4(PCLUCP)ct + λ 5(IRL)ct + λ 6(PCCDGE)ct+ ui (9) Where, STAFI is the food stability index and λ0 is the constant term; λ 1 to λ6 are regression coefficient for the


Kumar et al. (2015)

corresponding variables; and ui is the error term in Equation (9)(Kumar and Sharma, 2013): (ACCFI)ct = ξ0+ ξ1(PCAL)ct + ξ2(PCGDP)ct + ξ3(FP)ct + ξ4 (UR)ct + ξ5 (IMR)ct + ξ6 (LPR)ct + ξ7(PG)c t + ξ8(PGR)ct+ ui (10) Where, ACCFI is the food accessibility index and ξ0 is the constant term; ξ1 to ξ8 are the regression coefficient for the corresponding variables; and ui is the error term in Equation (10). In order to identify the causal relationship between poverty and GFSI, the following regression models were applied (Kumar and Sharma, 2013): (PG)ct = §0 + §1 (AVAFI)ct +§2 (STAFI)ct + §3 (ACCFI)ct + u i (11) Where, PG is the poverty gap; and AVAFI, STAFI, and ACCFI are the food availability index, food stability index and food accessibility index. §0 is the constant term and §1 to §3 are the regression coefficient for the corresponding variables; and ui is the error term as shown in Equation (11) (Kumar and Sharma, 2013): (PG)ct = €0 + €1 (GFSI)ct +u i

(12)

Where, PG and GFSI are the poverty gap and global food security index, respectively; €1 is regression coefficient for GFSI; and ui is the error term shown in Equation (12)(Kumar and Sharma, 2013): (GFSI)ct = θ0 + θ1 (PG)ct +u i

(13)

Where, GFSI and PG are the global food security index and poverty gap, respectively; θ0 is the constant term and θ1 is the regression coefficient for poverty gap; and ui is error term in Equation (13)(Kumar and Sharma, 2013). Selection of appropriate model All proposed regression models were run using STATA and SPSS statistical software. Several regressions were applied in order to select an appropriate model. Fixed effect and random effect models are used to identify the country effect and time effect in panel data (Demeke et al., 2011; Gupta et al., 2012; Kumar et al., 2015). Hausman specification test and Breusch-Pagan Lagrange Multiplier (LM) are used to check the quandary of fixed and random effect models (Gujarati, 2003; Kumar et al., 2015). Pesaran's test was used to determine the cross-

sectional independence. Modified Wald test is employed to address the group-wise heteroskedasticity (Gujarati, 2003; Kumar et al., 2015). Lagram-Multiplier test (Wooldridge test for autocorrelation) is used to identify the existence of serial-correlation/autocorrelation in panel data (Kumar et al., 2015). Finally, PCSEs and Driscoll-Kraay standard errors estimation models are applied in order to remove the presence of heteroskedasticity, serial-correlation, cross-sectional dependence and multicollinearity in all the proposed regression models (Gujarati, 2003; Kumar and Sharma, 2013; Kumar et al., 2015). RESULTS AND DISCUSSION Descriptions of the estimated global food security index (GFSI) for cross country Figures 1, 2, 3 and 4 show the trend in estimated global food security index (GFSI) for the selected set of economies in Europe and Central Asia, West and South East Asia, Latin American, and North and Sub-Sahara Africa. The estimated values of GFSI implies that Poland and Ukraine are most food secure countries in Europe and Central Asia region. Thailand and Malaysia are most food secure country, while, India and Pakistan are more food insecure country in West and South East Asian economies. In Latin American economies, Brazil and Mexico occupies a significant position in food security, whereas, Bolivia and Guatemala are food insecure country. In North and Sub-Sahara African region, Egypt and Morocco are better position in food security than other economies. The estimated values of GFSI also indicates that Ukraine, Hungry, Poland, Brazil and Romania are the most food secure countries among all the 31 cross economies and were ranked as 1st, 2nd, 3rd, 4th, 5th countries, respectively in the estimated GFSI in 2009(see Appendix: B). The aforementioned economies are high income and developed economies among all the 31 cross country. High cereal yield, per capita GDP, per capita land under cereal production and per capita arable land have increased food security in these regions. These countries also limit poverty, population growth rate, food inflation, depth of food-deficit and cereals imports dependency ratio. Developed economies are able to apply new technologies in cultivation to mitigate the negative effects of climate change. Further, in developed economies, high public spending in agricultural R&D has created more adaptable techniques to mitigate the harmful consequences of climate change in agricultural. While developing economies are unable


Kumar et al. (2015)

0.80 0.70 0.60 0.50 0.40

0.30 0.20

Hungary

Poland

Ukraine

Romania

Moldova

Kyrgyz Rep.

0.10

0.00 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Figure 1. Trend in estimat ed GF SI in Europe and Central Asian econo mies. Source: Author's estimation.

0.80

Turkey Sri Lanka

Malaysia Philippines

China Bangladesh

Thailand India

Indonesia Pakistan

0.60 0.40 0.20

0.00 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Figure 2. Trend in estimat ed GF SI in West and So uth East Asian eco nomies. Source: Author's estimation.

to adopt new technology in agricultural due to low per capita income and lack of financial resources. Another reason could be that most developed countries are highly industrious and as such, agriculture contributes less to their gross domestic product (GDP). Based on the aforementioned reasons, developed economies are efficient in providing ample food supply for their population. Upper income group countries like Turkey, Brazil, China, Mexico, Romania, Thailand, Turkey and Ukraine are also food secure country. Whereas, Colombia, Costa Rica, Ecuador, Malaysia and Panama are much lagged in GFSI. Rapid population growth and food price instability are major factors that contributes to food insecurity in lower and middle upper income economies. As such, lower middle and low income economies are unable to provide ample dietary energy supply to per person per day. Food security in most lower middle and low income countries are extremely alarming. However, Bangladesh, Egypt, Morocco, and

Kyrgyz Republic did well, as the values of GFSI for these countries have increased during 2000-2009(see Appendix: B). Sri Lanka, Honduras, Bolivia, India, Guatemala, Pakistan, Tanzania and Nigeria are the most insecure economies, they occupies 24th, 25th, 26th, 27th, 28th, 29th, 30th and 31st rank, respectively in the estimated GFSI in the year 2009 (see Appendix: B). There are many reasons that contributes to failure in achieving food security in these countries such as low cereal productivity and per capita availability of arable land, lowest cereal area and cereal production, minimal per capita GDP income and instability in food prices. Rapid population growth, increasing cereals imports dependency ratio and poverty trap also negatively affect food security in low middle and low income countries. These economies are located at lower latitude compared to high and developed economies. Therefore, climate change adversely affects agricultural production and food security in low middle and low income economies.


0.80 0.60

Kumar et al. (2015)

Mexico

Brazil

Costa Rica

Paraguay

Colombia

El Salvador

Ecuador

Honduras

Guatemala

Bolivia

Panama

0.40 0.20 0.00 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Figure 3. Trend in estimat ed GF SI in Latin A meric an econo mies . Sourc e: Aut hor's estimation.

0.50 0.40 0.30 0.20 0.10

Egypt

Morocco

Tanzania

Nigeria

0.00 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Figure 4. Trend in estimat ed GF SI in North and Sub Saharan A frican eco nomies. Source: Author's estimation.

Further, climate change brought several problems associated with agricultural productivity and food security in lower latitude countries, such as Bolivia, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, India, Indonesia, Malaysia, Moldova, Pakistan, Panama, Philippines, Sri Lanka, Tanzania, and Nigeria. India, Costa Rica, El-Salvador, Moldova, Morocco, Pakistan, Panama, Paraguay and Philippines are also in the most food insecure region. The estimated value of GFSI has declined in 2009 compared to previous report in 1991, 1995, 2000, and 2005 for corresponding economies (see Appendix: B). In summary, it can be said that due to high variability in socioeconomic and climatic factors, there exists a high food inequalities across world's major economies (Figure 1). India is the second most agriculture intensive country after China. Regardless of that, India is unable to feed

their population and as such, is known to have the largest number of hungry population with the highest number of undernourished children in the world. There are many reasons responsible for food insecurity in India and these are; low agricultural productivity, small land holdings, application of old technology in agricultural, financial constraints of farmers, low literacy of farmers, huge dependence of population on agricultural, high poverty trap, income inequality, distributional problems, regional disparities, rapid population growth and urbanization, low per capita income, high food price inflation, low public spending on agriculture R&D, and government inefficient policies (Figures 2 to 4). Also, several reasons are responsible for food insecurity in developing countries: low cereal productivity, low availability of per capita arable land,


cereal cultivated area and cereal production, lower per capita GDP and high variability in food prices. Rapid population growth rate, higher urbanization, and poverty trap are the major causes of food insecurity in lower middle and low income economies such as Bolivia, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, India, Indonesia, Malaysia, Moldova, Pakistan, Panama, Paraguay, Philippines, Sri Lanka, Tanzania and Nigeria. The increasing cereal imports dependency ratio is another reason for food insecurity in Bolivia, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, Malaysia, Panama, Philippines, and Nigeria. High inflation also a crucial reason for food insecurity in Bolivia, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, India, Indonesia, Malaysia, Moldova, Pakistan, Panama, Paraguay, Philippines, Sri Lanka, Tanzania, Nigeria. Empirical results and discussion Table 1 shows the association between GFSI and its components based on regression analysis. However, the results are not reliable and consistent due to high multicollinearity, therefore, it might be useful in the assessment of significant contribution of each component in GFSI. The estimates imply that all components are crucial to increase global food security. However, all components (that is, food availability, stability and accessibility) are positively associated with GFSI(Dev and Sharma, 2010; Joshi, 2012; Kumar and Sharma, 2013). The results also show that each components play a significant role to sustain food security at the global level. Table 2 reveals the regression results, which estimate the influence of socio-economic factors on the estimated GFSI. The empirical results indicate that arable land, cereal yield, forest area, and population engaged in agriculture, per capita land under cereal production, irrigated land, and per capita agriculture land are positively associated with GFSI. Here, it can be concluded that factor related to agricultural production activities plays a significant role in increasing global food security(Yu and You, 2013). Since food production is improved by increase in arable land, therefore, it would increase global food security. Cereal yield is the most important factor that increase global food security (Salami, 2011; Prabha et al., 2010). Forest area could be crucial exogenous factor to increase agricultural productivity and food security, since the area can maintain environmental sustainability and mitigate the negative consequences of climate change in relation to

Kumar et al. (2015)

activities of cultivation. Further, it could be an ecosystem-based adaptation technique to mitigate the adverse effect of carbon emission in cultivation (Pramova et al., 2012), and to increase food production and global food security. Creating more employment opportunities for people in the agriculture sector could be beneficial in two ways: first, it may increase agricultural productivity, and second, it may increase the economic capacity of the working population. Moreover, irrigated area has higher productivity than non-irrigated area (Kumar et al., 2015). Therefore, irrigated area may also increase food production and global food security. In addition, per capita arable land is a crucial variable that increase global food security. Per capita electric consumption, energy use, and dietary energy supply also play a positive and significant role in increasing global food security. Since, these variables are positively associated with food security and human health. Per capita food production variability and depth of fooddeficit have negative and statistically, significant impacts on global food security. This shows that variability in food production may lead to increase in the price of food and can also reduce the economic capacity of a population (Pandey, 2009), consequently resulting in decline of global food security. Infant mortality rate, poverty gap and population growth rate also have a negative and statistically, significant impact on global food security. Inappropriate health facilities are the major cause of high infant mortality rate, which increases the burden on poor population. Swaminathan (1998) and Kumar and Sharma (2013) observed that high infant mortality rate and poverty are harmful factors of food security in India. High population growth increases the demand for food products in domestic market. Additional food demand can increase food prices and decrease the economic capacity of people to acquire food (Clark et al., 2010). Female participation, urbanization, and labour participation have positive and significant relationship with global food security. Urbanization may create employment opportunities for people and increase economic capacity of a population. Therefore, people would be in position to acquire food according to their needs, which enhance food security. However, several studies have shown that urbanization has negative influence on food security in most developing economies (Mallet al., 2006; Johnson, 2009; Joshi, 2012). Here, it can be argued that urbanization can play positive role in increasing food security for countries that are able to maintain the pace of urban growth rates (EIU, 2012, 2013, 2014, 2015).1 Per capita carbon emission have negative impact on global food security. This implies that increase in the quantity of


Kumar et al. (2015)

Table 1. Regressio n result based on linear r egressio n, correlated panels correct ed st andard errors (PCSEs) model for GF SI.

No. of Observations No. of Countries No. of Obs./Countries Variable AVAF STAF ACCF Con. Coef.

Reg. Coe. 0.31134* 0.31042* 0.36379* 0.00723*

620 31 20

R-squared Wald chi2 Prob > chi2

Panel Corr. Std. errors 0.00179 0.00341 0.00204 0.00201

z 173.72 90.94 178.21 3.60

0.9997 142389.54 0.0000 P > |z| 0.000 0.000 0.000 0.000

95% confidence interval 0.30783 0.31485 0.30373 0.31711 0.35979 0.36779 0.00329 0.01116

Source: Author's estimation.

Table 2. Regressio n result based on linear r egressio n, correlated panels correct ed st andard errors (PCSEs) model for GF SI.

No. of Observations No. of Countries No. of Obs./Countries

617 31 20

R-squared Wald chi2 (20) Prob > chi2

Variable

Panel Corr. Std. errors

z

ARL PCEPC PCEU PCFPV PCDES PCDFD CY FA EAGS PCLUCP IRL PCCDGE PCAL PCGDP FP UR IMR LPR PG PGR Con. Coef.

Reg. coefficient 0.00085* 0.00002* 0.00001* -0.00078* 0.00004* -0.00015* 7.82e-06* 0.00099* 0.00040* 0.27288* 0.00049* -0.00269*** 0.04572* 7.47e-07 0.00770* 0.01290* -0.00033* 0.00101* -0.00071* -0.00620* -0.25233*

0.00004 2.01e-06 4.18e-06 0.00008 4.28e-06 0.00002 6.24e-07 0.00005 0.00007 0.02720 0.00003 0.00140 0.00830 1.23e-06 0.00051 0.00126 0.00007 0.00009 0.00010 0.00135 0.02447

0.9659 46152.12 0.0000

P > |z| 21.53 8.73 2.86 -9.52 10.01 -9.03 12.54 18.68 5.72 10.03 17.06 -1.92 5.51 0.61 15.21 10.25 -5.00 11.55 -7.41 -4.62 -10.31

0.000 0.000 0.004 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.055 0.000 0.544 0.000 0.000 0.000 0.000 0.000 0.000 0.000

95% confidence interval 0.00077 0.00001 3.76e-06 -0.00094 0.00003 -0.00018 6.60e-06 0.00089 0.00026 0.21958 0.00044 -0.00543 0.02945 -1.66e-06 0.00671 0.01044 -0.00045 0.00084 -0.0009 -0.00883 -0.30029

0.00092 0.00002 0.000020 -0.00062 0.00005 -0.00011 9.04e-06 0.00109 0.00054 0.32618 0.00055 0.00005 0.06198 3.16e-06 0.00870 0.01537 -0.00020 0.00119 -0.00052 -0.00357 -0.20437


carbon emission could cause more sensitivity in climatic factors. Hence, agricultural productivity and global food security could be negatively impacted as increase in per capita carbon emission (Zhang, 2008; Gadgil, 1995; Girardet and Bree, 2009; Mall et al., 2006).

Table 3 shows the regression results, which assess the impact of the undertaken factors on food availability. Arable land, per capita electric power consumption, energy use and dietary energy supply have positive and significant association with food availability. Per capita


Kumar et al. (2015)

Table 3. Regressio n result with linear regression, correlated panels corrected standard errors (PCSEs) model for food availability.

No. of Observations No. of Countries No. of Obs./Countries Variable ARL PCEPC PCEU PCFPV PCDES PCDFD Con. Coef.

Reg. coefficient 0.00272* 0.00005* 0.00005* -0.00299* 0.00011* -0.00060* 0.09276*

617 31 19


Panel Corr. Std. errors 0.00005 3.47e-06 5.06e-06 0.00018 3.94e-06 0.00002 0.01123

z 53.48 14.77 8.80 -17.13 27.94 -32.68 8.2600

0.9790 6641.05 0.0000 P > |z| 0.000 0.000 0.000 0.000 0.000 0.000 0.000

95% confidence interval 0.00262 0.00282 0.00004 0.00006 0.00003 0.00005 -0.00333 -0.00265 0.00010 0.00012 -0.00064 -0.00057 0.07074 0.11477


Table 4. Regressio n result based on linear r egressio n, correlated panels correct ed st andard errors (PCSEs) model for food stability.

No. of Observations No. of Countries No. of obs./Countries Variable CY FA EAGS PCLUCP IRL PCCDGE Con. Coef.

Reg. coefficient 0.00003* 0.002434* 0.00221* 0.59068* 0.00168* -0.01786* 0.13012*

620 31 20 Panel Corr. Std. errors 8.31e-07 0.00001 0.00006 0.02078 0.00004 0.00066 0.00511

R-squared Wald chi2 Prob > chi2 z 31.33 40.83 35.15 28.43 39.83 -27.17 25.47

0.9556 8843.35 0.0000 P > |z| 0.000 0.000 0.000 0.000 0.000 0.000 0.000

95% confidence interval 0.00002 0.00002 0.00231 0.00257 0.00209 0.00233 0.54996 0.63141 0.00160 0.00177 -0.01914 -0.01656 0.12010 0.14013


electric power consumption, energy use and dietary energy supply are also positively related with food availability. Estimates specify that an increase in these factors could be useful to increase global food security and health of population. Per capita food production variability and per capita depth of food-deficit have negative and statistically, significant impact on food availability. This could be attributed to the fact that food variability increases the price of food and reduce the economic capacity of population to buy food from domestic market. The depth of food-deficit is also a harmful factor that decreases the efficiency of a given population. Table 4 shows the estimate of the impact of explanatory factors on food stability. The empirical results shows that all agriculture production related

factors has a positive and statistically, significant impact on food stability. Food stability is positively associated with cereal production. This means that increase in the yield cereal can improve food stability (Salami, 2011). Forest area could be a better solution to increase food stability. Participation of agricultural labour in agriculture activities and per capita land under cereal production have positive impact. Irrigated area is a vital factor to improve food stability, since irrigated area has higher yielding capacity than non-irrigated area (Mondal et al., 2014; Kumar et al., 2015). Therefore, agricultural productivity and food stability can be improved by the extension of irrigated area under cultivation. Per capita carbon emission has a negative and statistically significant effect on food stability. This could be attributed to the fact that increase in carbon emission


Kumar et al. (2015)

Table 5. Regressio n result based on linear r egressio n, correlated panels correct ed st andard errors (PCSEs) model for food acc essibility.

No. of Observations No. of Countries No. of Obs./countries Variable PCAL PCGDP FP UR IMR LPR PG PGR Con. Coef.

Reg. coefficient 0.24133* 7.58e-06* 0.02339* 0.03890* -0.00123* 0.00258* -0.00234* -0.02270* -0.90312*

619 31 20


Panel Corr. Std. errors 0.02763 2.47e-06 0.00414 0.00531 0.00029 0.00049 0.00042 0.00539 0.21587

z 8.74 3.07 5.65 7.33 -4.29 5.23 -5.56 -4.21 -4.18

0.9329 2008.89 0.0000 P > |z| 0.000 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000

95% confidence interval 0.18718 0.29548 2.74e-06 0.00001 0.01528 0.03150 0.02849 0.04930 -0.00179 -0.00067 0.00161 0.00355 -0.00316 -0.00151 -0.03325 -0.01214 -1.32620 -0.48002


Table 6. Regressio n result based on Driscoll-Kraay standard errors model for co mponents o f GFSI and pov erty gap.

No. of observations No. of countries No. of obs./countries Variable AVAF STAF ACCF Con. Coef.

Reg. coefficient -7.32516* 14.49743 -71.07993* 43.80061*

620 31 20 Panel Corr. Std. errors 13.00446 29.82621 13.2322 16.33807

R-squared F-value Prob > F z -0.56 0.49 -5.37 2.68

0.5231 18.96 0.0000 P > |z| 0.577 0.630 0.000 0.012

95% confidence interval -33.8838 19.23348 -46.4158 75.41067 -98.1037 -44.0562 10.4338 77.1674


creates foundation for climate change (Gadgil, 1995; Mall et al., 2006; Zhang, 2008; Girardet and Bree, 2009). Hence, climate change has negative impact on agricultural productivity and food stability. Table 5 reveals the impact of several variables on food accessibility. Estimates have shown that per capita arable land, per capita GDP and labor participation rate have positive and statistically, significant impact on food accessibility. This implies that food accessibility will increase by increases in per capita arable land and labor participation in economic activities. Per capita GDP is the most significant factor of improving food accessibility. It shows that food accessibility will improve as the income of the people is increased. Female population and urbanization are positively associated with food accessibility. Infant mortality rate, poverty gap and population growth rate are detrimental factors of food

accessibility. Here, it can be argued that high infant mortality rate can produce several obstacles to the individual. The estimate of study also shows that poverty has negative association with food accessibility. Rapid population growth rate has negative implications on food accessibility, since it increases food demand in domestic market and increase food prices. Table 6 shows the empirical results, which measure the association of poverty gap with each components of global food security. Food availability and accessibility are negatively associated with poverty gap. This relationship suggest that food availability and accessibility play significant role in poverty eradication. It also shows that food accessibility would be useful to trap out people from poverty. On the contrary, food stability is positively associated with poverty and it implies that increase in food stability only, would not be


Kumar et al. (2015)

Table 7. Regressio n result based on Driscoll-Kraay standard errors model for GFSI and pov erty gap.

No. of observations No. of countries No. of obs./countries Variable GFSI Con. Coef.

620 31 20

Reg. coefficient -123.13130* 69.98715*

Panel Corr. Std. errors 19.42975 10.03454

R-squared F-value Prob > F z -6.34 6.97

0.4221 40.16 0.0000 P > |z| 0.000 0.000

95% confidence interval -162.812 -83.4504 49.49388 90.48042


Table 8. Regressio n result based on Driscoll-Kraay standard errors model for pov erty gap and GFSI.

No. of observations No. of countries No. of obs./countries

620 31 20

R-squared F-value Prob > F

0.4221 33.43 33.43

Variable

Reg. coefficient

Panel Corr. Std. errors

z

P > |z|

PG Con. Coef.

-0.00343* 0.49647*

0.00060 0.01636

-5.78 30.34

0.000 0.000

95% confidence interval -0.00464 0.46306

-0.00222 0.52989


helpful in poverty eradication. Hence, there must be consistent equilibrium among all components of food security from national to global level. Tables 7 and 8 demonstrate the regression results, which investigate the casual relationship between GFSI and poverty gap. Estimates show that GFSI is negatively associated with poverty. This means that increase in global food security would be crucial for poverty eradication. Poverty and global food security have a causal relationship with each other. This implies that increase in poverty would create obstacles in sustaining global food security (Kramer, 2007). Kumar and Sharma (2013) also observed a causal relationship between poverty and food security, and vice-versa in India. CONCLUSION AND POLICY IMPLICATIONS The present study was carried to investigate the impact of carbon emission and other socio-economic factors on global food security. In this study, global food security index (GFSI) was generated using descriptive method with its three components for 31 cross country during 1990-2009 (see Appendix: A). Descriptive results show that developed economies are in better position than large agrarian and agriculture intensive countries. India, Pakistan, and Nigeria are agrarian economies despite

their being ranked 27th, 31st and 29th, respectively across 31 countries (see Appendix: B). There is exists a high food inequalities across regions and economies in the world. Empirical results imply that all components of global food security, such as food availability, stability and accessibility, play a crucial role in sustaining food security at the global level. Per capita carbon dioxide gas emission is negatively associated with food security. Therefore, world's economies are required to abate carbon emissions using alternative techniques of production in order to ensure food security. Further, it has been estimated that food security could be useful in poverty eradication (Kramer, 2007). Poverty and global food security have causal relationship (Hollaender, 2010; ADB, 2012). The creation of more employment opportunities for agricultural labour in non-agricultural sector may be crucial to sustain food security. High population growth and infant mortality rate, and food production variability rate have negative effect on global food security. This study provides several policy suggestions to sustain food security. Global policy maker needs to reduce carbon emission. International policy makers and development thinkers also need to adopt conductive plan for poverty eradication, and it would be the most important component to avoid food insecurity and chronic hunger. Food insecurity and poverty also


coincide with high incidence of high infant mortality rate. Increase in cereal yield, irrigated area, area under cereal production, and forest area could be better solution in ensuring global food security. By effecting regulation to control high population growth and infant mortality rate, variability in food production may be useful in the improvement of global food security. In order to achieve food security, developing and large agrarian economies need to apply favorable land and water conditions that are needed in order to exploit their potential to increase agricultural productivity through a more conductive policy framework and increased investment in agricultural and rural development (Yu and You, 2013; Kumar et al., 2015). Poverty eradication, price stability in food grain product, irrigation facilities in agriculture, development of infrastructure, increasing per capita income from non-agriculture sector, food production, and adoption of modern technology in agriculture are useful in ensuring food security at national and global level. The empirical findings of the present study was based on thirty one cross country of the world, therefore, statistical inference may not be generalized for any economy. Since, all economies have high variation in socioeconomic variables, policy factors, political environment and climatic factors, this study could not investigate the influence of income distribution, food consumption expenditure, income inequalities, food expenditure, and policy factors on food security within specific country. Furthermore, this stable was able assess the food inequalities as a global food security index across selected set of economies. Therefore, the estimated values of GFSI of a specific country cannot be consistent with the findings of existing studies, which incorporated more than 31 cross country (EIU, 2012, 2013, 2014, and 2015 1; Yu and You, 2013). Hence, further research on a specific region/country is required in order to determine the reliability and consistency of the empirical results of the present study. ACKNOWLEDGEMENTS The study is an output of a scholarship from the Food Security Center, University of Hohenheim, Stuttgart (Germany) which is part of the DAAD (German Academic Exchange Service) program “exceed” and is supported by DAAD and the German Federal Ministry for Economic Cooperation and Development (BMZ). This study was also supported by Asian Institute of Technology, Bangkok (Thailand) as a host institute and the Indian Institute of Technology Indore (Madhya Pradesh) India

Kumar et al. (2015)

as a co-funding organization. This study is based on PhD research project of the first author. The author is also grateful to the Ministry of Human Resource Development (GoI) for providing scholarship in pursuing PhD (Economics) at School of Humanities and Social Sciences, IIT Indore. REFERENCES Aggarwal, K. P. (2008). Climate change and its impact on agriculture and food security. LEISA, Bangalore: India. Ajetomobi, J., Abiodun, A., and Hassan, R. (2011). Impacts of climatechange on rice agriculture in Nigeria.Trop. Subtrop. Agro-ecosyst.,14(2): 613-622. Alam Md. Q. (2013). Climate change, agriculture productivity and economic growth in India: The bound test analysis. Int. J. Appl. Res. Stud., 2(11): 1-14. Arndt, C., Chinowsky, P., Sherman, R., Kenneth, S., Finn, T., and Thurlow, J. (2012). Economic development under climate change. Rev. Dev. Econ.,16(3): 369-377. Asian Development Bank (ADB) (2012). Food security and poverty in Asia and the pacific: Key challenge and policy issues. Mandaluyong City: Philippines. Burke, M., and Lobell, D. (2010). (Chapter 2: Climate effects on food security: An overview) Climate Change and Food Security. Adv. Glob. Change Res., 37: 13-30. Clark, W. C., Kristjanson, P., Campbell, B., Juma, C., Holbrook, N. M., Nelson, G., and Dickson, N. (2010). Enhancing food security in an era of global climate change. Center for International Development (CID), Harvard University, Working paper 198. Demeke, A. B., Keil, A., and Zeller, M. (2011). Using panel data to estimate the effect of rainfall shock on smallholders food security and vulnerability in rural Ethiopia. Clim. Change, 108: 185-206. Deressa, T., Hassan, R., and Poonyth, D. (2005). Measuring the impact of climate change on South African agriculture: The case of sugarcane growing regions. Agrekon, 44(4): 524-542. Dev, S. M., and Sharma, A. N. (2010). Food security in India: Performance, challenges and policies. Oxfam India Working Paper 07. Food and Agriculture Organization of the United Nations (FAO) (2002). The state of food insecurity 2001. Rome: Italy. Food and Agriculture Organization of the United Nations (FAO) (2009). Food security and agricultural mitigation in developing countries: options for capturing synergies. Rome: Italy. Food and Agriculture Organization of the United Nations (FAO) (2008). Climate change and food security: a


framework document. Rome: Italy. Gadgil, S. (1995). Climate change and agriculture-an Indian perspective. Curr. Sci., 69(8): 649-659. Gbetibouo, G. A., and Hassan, R. M. (2005). Measuring the economic impact of climate change on major South African field crops: A Ricardian approach. Global and Planet. Change, 47: 143-152. Girardet, H., and Bree, A. (2009). Cultivating the future: Food in the age of climate change. World Future Council Bei den Muhren, Hamburg: Germany. Greg, E. E., Anam, B. E., William, M. F., and Duru, E. J. C. (2011). Climate change, food security and agricultural productivity in African: Issues and policy directions. Int. J. Human. Soc. Sci.,1(21): 205-223. Gregory, P. J., Ingram, J. S. I., and Brklacich, M. (2012). Climate change and food security. Philosophical Trans. Royal Soc. B, 360: 2139-2148. Gujarati, D. (2003). Basic Econometrics. New York. McGraw Hill. Gupta, S., Sen, P., and Srinivasan, S. (2012). Impact of climate change on Indian economy: Evidence from food grain yields. Centre for Development Economics Working Paper 218. Hoddinott, J. (1999). Choosing outcome indicators of household food security: Technical guide. International Food Policy Research Institute (IFPRI), Washington D.C., U.S.A. Hollaender, M. (2010). Human right to adequate food: NGOs have to make the difference. CATALYST Newsletter of Cyriac Elias Voluntary Association (CEVA), 8(1): 5-6. Horowitz, J. K. (2009). The income-temperature relationship in a cross-section of countries and implications for predicting the effect of global warming. Env. Resour. Econ., 44: 475-493. Ibrahim, H., Bello, M., and Ibrahim, H. (2009). Food security and resource allocation among farming households in north central Nigeria. Pakistan J. Nutr., 8(8): 1235-1239. Johnson, R. (2009). Food security: The role of agricultural trade. International Food and Agricultural Trade Policy Council, IPC Discussion Paper 2009, Washington D.C., U.S.A. Joshi, L. (2012). Climate change and food security: A micro study of drivers and impacts in western Maharashtra, India. WOTR, Pune: India. Julia, M. S., Challinor, A. J., Hoskins, B. J., and Timothy, R. W. (2005). Introduction: Food crops in a changing climate. Philosophical Trans. Royal Soc. B, 360: 1983– 1989. Kramer, A. M. (2007). Adaptation to climate change in poverty reduction strategies. Human Development

Kumar et al. (2015)

Report Office Occasional Paper 34. Kumar, A., and Sharma, P. (2013). Impact of climate variation on agricultural productivity and food security in rural India. Economics Discussion Papers 2013-43, Kiel Institute for the World Economy. Kumar, A., Sharma, P. and Ambrammal S. K. (2015). Climatic effects on sugarcane productivity in India: a stochastic production function application. Int. J. Economics and Business Research, 10(2), 179–203. Kurukulasuriya, P., and Mendelsohn, R. (2008). A Ricardian analysis of the impact of climatechange on African cropland (Special issue: Climatechange and African agriculture). Afr. J. Agric. Resour. Econ., 2(1): 123. Lee, H.-L. (2009). The impact of climate change on global food supply and demand food prices, and land use. Paddy Water Env., 7: 321-331. Mall, M. K., Singh, R., Gupta, A., Srinivasan, G., and Rathore, L. S. (2006). Impact of climate change on Indian agriculture: A review. Clim. Change, 78: 445478. Masters, G., Baker, P. and Flood, J. (2010). Climate change and agricultural commodities. CABI Working Paper 02(38). Mendelsohn, R., Massetti, E., and Kim, C. G. (2011). The impact of climate change on US agriculture. J. Rural Dev. (Seoul), 34(2): 19-43. Mendelsohn, R., Nordhaus, W. D., and Shaw, D. (1994). The impact of global warming: A Ricardian analysis. Am. Econ. Rev., 84(4): 753-771. Mondal P., Jain M., Robertson A. W., Galford G. L., Mall C., and DeFries R. S. (2014). Winter crop sensitivity to inter-annual climate variability in central India. Clim. Change, 126: 61-76. DOI 10.1007/s10584-014-1216-y Nath, P. K., and Behera, B. (2011). A critical review of impact and adaptation to climate change in developed and developing countries. Env. Dev. Sustain., 13: 141162. Omotesho, O. A., Adewumi, A., Mahammad-Lawal, A., and Ayinde, O. E. (2006). Determinants of food security among the rural farming households in Kwara State, Nigeria. Afr. J. Gen. Agric., 2(1): 7-15. Pandey, S. (2009). Adaptation to and mitigation of climate change in agriculture in developing countries. IOP Conference Series: Earth and Environmental Science, p. 8. Pant, K. P. (2009). Effect of agriculture on climate change: A cross country study of factors affecting carbon emissions. J. Agric. Env., 10: 72-88. Prabha, Goswami, K., and Chatterjee, B. (2010). Linkage between rural poverty and agricultural productivity across the districts of Uttar Pradesh in India. J. Dev.


Agric. Econ., 2(2): 026-040. Pramova, E., Locatelli, B., Djoudi, H., and Somorin, O. A. (2012). Forests and trees for social adaptation to climate variability and change. WIREs Clim., 3: 581596. Ramasamy, J., and Moorthy, P. (2012). Managing food insecurity and poverty in India in the era of globalization. Int. J. Multidisciplinary Res., 2(1): 411421. Ranganathan, C. R., Palanisami, K., Kakumanu, K. R., and Baulraj, A. (2010). Mainstreaming the adaptations and reducing the vulnerability of the poor due to climate change. ADBI Working Paper Series 333, Rome: Italy. Ranuzzi, A., and Srivastava, R. (2012). Impact of climate change on agriculture and food security. ICRIER Policy Series 16. Rukhsana (2011). Dimension of food security in a selected state Uttar Pradesh. J. Agric. Extension Rural Dev., 3(2): 29-41. Sajjad H., Nasreen I., and Ansari S.A. (2014). Assessment of spatio-temporal dynamics of food security based on food security index analysis: A case from Vaishali district, India. J. Agric. Sustain., 5(2): 125-152. Salami, A., Brixiova, Z., Kandil, H., and Mafusire, A. (2011). Towards food security in Africa: Challenges, policies and the role of the African development bank. Africa Econ. Brief, 2(2). Schmidhuber, J., and Tubiello, F. N. (2007). Global food security under climate change. Proceedings of the National Academy of Sciences of the United States of America (PNAS), 104(50). Seo, S. N., and Mendelsohn, R. (2007). A Ricardian analysis of the impact of climatechange on Latin American farms.World Bank Policy Research Working Paper 4163. Seo, S. N., and Mendelsohn, R. (2008). A Ricardian analysis of the impact of climate changes on South American farms. Chilean J. Agric. Res., 68(1): 68-79.

Kumar et al. (2015)

Shakeel, A., Jamal, A., and Zaidy, N. (2012). A regional analysis of food security in Bundelkhand Region (Uttar Pradesh, India). J. Geogr. Regional Plan., 5(9): 252-262. Swaminathan, M. S. (1998). Population, environment and food security. J. Indian Soc. Agric. Statist., 51(2-3): 99112. Ye L., Xiong W., Li Z., Yang P., Wu W., Yang G., Fu Y., Zou J., Chen Z., Ranst E.V, and Tang H. (2013) Climate change impact on China food security in 2050. Agron. Sustain. Dev., 33: 363-374. Yu B., and You, L. (2013). A typology of food security in developing countries. China Agric. Econ. Rev., 5(1): 118-153. Yu, B., Zhu, T., and Hai, N. M. (2010). Impacts of climate change on agriculture and policy options for adaption: The case of Vietnam. IFPRI Discussion Paper 01015. Zhai, F., and Zhuang, J. (2009). Agriculture impact of climate change: A general equilibrium analysis with special reference to Southeast Asia. ADBI Working Paper 131. Tokyo: Asian Development Bank Institute. Zhai, F., Lin, T., and Byambadori, E. (2009). A general equilibrium analysis of the impact of climate change on agriculture in the people's republic of China. Asian Dev. Rev., 26(1): 206-225. Zhang, Y., Shaohong, W. U., Erfu, D. A. I., Dengwei, L. I. U., and Yune, Y. I. N. (2008). Identification and categorization of climate change risks. Chinese Geogr. Sci., 18(3): 268-275. 1http://www.foodsecurityindex.eiu.com.


Kumar et al. (2015)

APPENDIX Appendix A: Group of cross-countries (World Bank, 2011): High income countries: Hungary, and Poland. Upper middle income countries: Brazil, China, Colombia, Costa Rica, Ecuador, Malaysia, Mexico, Panama, Romania, Thailand, and Turkey. Lower middle income countries: Bolivia, Egypt, El Salvador, Guatemala, Honduras, India, Indonesia, Moldova, Morocco, Pakistan, Paraguay, Philippines, Sri Lanka, Ukraine, and Nigeria . Low income countries: Bangladesh, Kyrgyz Republic, and Tanzania. Appendix B Appendix B: Estimated value of glo bal food sec urity index (GFSI) for various countries .

Country /Year Hungary Poland Ukraine Romania Moldova Turkey Mexico Brazil Thailand Kyrgyz Rep. Costa Rica Malaysia Paraguay Colombia Egypt Panama China Morocco El Salvador Indonesia Ecuador Sri Lanka Philippines Honduras Tanzania Bangladesh Guatemala Bolivia India Nigeria Pakistan

1990 0.675 (1) 0.654 (2) 0.617(3) 0.562 (4) 0.536 (5) 0.523 (6) 0.508 (7) 0.504 (8) 0.504 (9) 0.481 (10) 0.473 (11) 0.470 (12) 0.447 (13) 0.438 (14) 0.435 (15) 0.430 (16) 0.428 (17) 0.417 (18) 0.416 (19) 0.413 (20) 0.412 (21) 0.410 (22) 0.409 (23) 0.405 (24) 0.389 (25) 0.383 (26) 0.379 (27) 0.376 (28) 0.374 (29) 0.334 (30) 0.320 (31)

1995 0.632 (2) 0.625 (3) 0.674 (1) 0.570 (5) 0.574 (4) 0.532 (8) 0.540 (6) 0.516 (9) 0.536 (7) 0.510 (10) 0.509 (11) 0.504 (12) 0.462 (14) 0.471 (13) 0.457 (15) 0.450 (17) 0.452 (16) 0.444 (21) 0.450 (18) 0.446 (20) 0.448 (19) 0.421 (24) 0.426 (22) 0.423 (23) 0.358 (30) 0.394 (26) 0.408 (25) 0.382 (28) 0.390 (27) 0.362 (29) 0.349 (31)

2000 0.622 (3) 0.625 (2) 0.648 (1) 0.552 (4) 0.532 (7) 0.481 (12) 0.519 (8) 0.545 (6) 0.550 (5) 0.491 (10) 0.495 (9) 0.491 (11) 0.449 (18) 0.471 (13) 0.450 (17) 0.458 (16) 0.468 (15) 0.408 (23) 0.470 (14) 0.428 (19) 0.393 (27) 0.419 (21) 0.409 (22) 0.403 (25) 0.351 (31) 0.425 (20) 0.398 (26) 0.405 (24) 0.390 (28) 0.358 (30) 0.358 (29)

Source: Author's estimation. Values in bracket represents the rank for respective countries.

2005 0.627 (1) 0.612 (2) 0.609 (3) 0.569 (4) 0.507 (9) 0.514 (8) 0.526 (7) 0.551 (6) 0.563 (5) 0.477 (12) 0.472 (13) 0.497 (10) 0.456 (17) 0.479 (11) 0.443 (21) 0.455 (19) 0.456 (18) 0.458 (16) 0.458 (15) 0.446 (20) 0.467 (14) 0.403 (24) 0.412 (23) 0.399 (25) 0.350 (29) 0.412 (22) 0.363 (28) 0.388 (26) 0.378 (27) 0.315 (31) 0.346 (30)

2009 0.643 (2) 0.639 (3) 0.664 (1) 0.584 (5) 0.515 (10) 0.525 (8) 0.531 (7) 0.591 (4) 0.577 (6) 0.498 (12) 0.479 (14) 0.521 (9) 0.453 (20) 0.499 (11) 0.447 (21) 0.467 (17) 0.493 (13) 0.456 (19) 0.474 (15) 0.466 (18) 0.471 (16) 0.411 (25) 0.426 (23) 0.421 (24) 0.341 (30) 0.444 (22) 0.384 (28) 0.403 (26) 0.385 (27) 0.330 (31) 0.345 (29)