Full length article
Rural Sudanese household food consumption patterns
Raga Elzaki
a,b, Muhammet Yunus Sisman
c, Mohammed Al-Mahish
a,⇑a
Department of Agribusiness and Consumer Science, College of Agriculture and Food Science, King Faisal University, AlAhsa 31982, Saudi Arabia
bDepartment of Rural Economics and Development, Faculty of Animal Production, University of Gezira, Sudan c
Dumlupınar University, Faculty of Economics and Administrative Sciences, Kütahya, Turkey
a r t i c l e i n f o
Article history: Received 17 June 2020 Revised 14 August 2020 Accepted 16 November 2020 Available online xxxx Keywords:Quadratic Almost Ideal Demand System (QUAIDS)
Food consumption Food demand elasticities Demographic variables
a b s t r a c t
The paper aims to estimate Sudanese household food demand by focusing on six aggregated food items. The items are cereals (sorghum, millet, wheat), meat and chicken, milk and eggs, vegetables (okra, onion, tomato), staples (sugar, salt, oil), and caffeine intake (tea and coffee). The paper used the Quadratic Almost Ideal Demand System (QUAIDS) model to estimate Sudanese household food demand. The results showed that demographical variables have an impact on households’ expenditure shares. For example, the results revealed that married respondents spend less from their income on cereals and staples com-pared to unmarried respondents. Also, men spend more from their income on caffeine comcom-pared to women. The results of the Marshallian own-price elasticities showed that all food items in Sudan are price inelastic indicating that price changes have a small impact on quantity demanded. Also, expendi-ture elasticities show that all food items in Sudan classified as necessities, except for meat and chicken. Ó 2020 The Authors. Production and hosting by Elsevier B.V. on behalf of King Saud University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction
The general feature of Sudan’s topography is fertile, agricultural land (Mahgoub, 2014) that is suitable for growing most food crops, that is, instant cereal crops (sorghum, wheat and millets), which dominate crop production in Sudan and provide nearly 53% of the population’s daily calorie requirements (FAO-SIFSIA, 2010; Ibnouf, 2011). Due to wide variety of crops, there is no single crop which dominates the national food basket, the sorghum crop is considered to be the main food basket in central and northern regions, while the millet crops represent the main food basket in western regions. Currently, Sudan is self-sufficient in traditional cereal crops, such as sorghum and millet, which are considering to be the main staple crops in the majority of the Sudanese popu-lation (FAO, 2019a). Sorghum and millet are usually found in unfa-vourable environments where many poor farmers exist (Alwang et al., 2017).
Sudan’s population was estimated to be more than 40 million people in 2017 and this figure is expected to increase annually by 2.42% (World Bank, 2019). Food consumption patterns have changed as a result of increase in income and adverse rural–urban migration movement. Previous studies conducted on food sumption pattern using Angel curve showed that the food con-sumption behaviours of households at different positions on the social ladder are different in Sudan (Hassan and Babu, 1991). By comparing food consumption, in the form of dietary energy con-sumption (kcal/person/day) in some certain African countries, Fig. 1 shows the dynamic trend in food consumption in certain African countries.Fig. 1suggests that Egypt recorded the highest food consumption throughout the years, whereas Eretria displays less food consumption during the same periods. In case of Sudan, the food consumption starts to increase gradually during period of 2000–2018 and reaches the highest consumption points in peri-ods of 2016–2018. Likewise, it is noticed that the food consump-tion in Africa, generally, was less during 2006–2008, and this might be related to the global food prices swing during this period (FAO, 2009). Food consumption in African developing countries is forecasted to be 3050 kcal per capita per day in 2030 (Vasileska and Rechkoska, 2012).
To the best of the authors’ knowledge, no recent study has anal-ysed food consumption in Sudan, rather, most studies focus on estimating poverty incidences, food security, malnutrition and bal-anced diet (Musaiger, 1993; Ali Taha, 1978; Von Braun and Zaki, 1992; Hassan and Babu, 1991, Ibnouf, 2009; Khalid et al., 2017).
https://doi.org/10.1016/j.jssas.2020.11.004
1658-077X/Ó 2020 The Authors. Production and hosting by Elsevier B.V. on behalf of King Saud University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
⇑Corresponding author.
E-mail address:malmahish@kfu.edu.sa(M. Al-Mahish). Peer review under responsibility of King Saud University.
Production and hosting by Elsevier
Contents lists available atScienceDirect
Journal of the Saudi Society of Agricultural Sciences
j o u r n a l h o m e p a g e : w w w . s c i e n c e d i r e c t . c o m
Please cite this article as: R. Elzaki, M. Yunus Sisman and M. Al-Mahish, Rural Sudanese household food consumption patterns, Journal of the Saudi Society of Agricultural Sciences,https://doi.org/10.1016/j.jssas.2020.11.004
Accordingly, this paper is the first article that uses Quadratic Almost Ideal Demand System (QUAIDS) to analyse household food consumption in Sudan. This study incorporates household demo-graphic variables, which are supposed to be chief factors influenc-ing households’ consumption. Moreover, economic distortion, price fluctuations, drought, civil wars and/or conflicts, economic sanctions and income inequality among households have agitated food production and consumption in Sudan (Siddig, 2010). More-over, at the international level, few papers have analysed the changes in consumer demand or expenditure shares in the estima-tion of demand elasticities (Shashika et al., 2019). The paper aims to estimate income elasticities, food own-price elasticity, and food cross-price elasticities in Sudan. Also, the paper examines the effects of household demographical characteristics on households’ food expenditures.
2. Literature review
The food consumption patterns analysis has taken more atten-tion of naatten-tions and authors since it reflects on consumers’ beha-viour. The income expenditure of food durable consumption goods and human capital (education, health, marriage, etc.) are treated as investment expenditures (Hassan and Babu, 1991). Therefore, there has been a growing interest in investigating food consumption studies, particularly in developing countries where food consumption continues to be a principal objective of policy initiatives in low-income countries (Delgado, 2003; Kearney, 2010; Zezza et al., 2017). Heterogeneities in food consumption may be significant across households because of dissimilarities in preferences, food prices, food availabilities and resource con-straints (Humphries et al., 2017).
Li and Yu (2010)confirmed that income, education, and house-hold stock raising have substantial impact on food consumption. Difference in food consumption satisfaction is indistinctively con-nected to characteristics such as per capita income, land, and non-farm employment. Likewise, the rising of food prices affects and/or determines food consumption (Green et al., 2013). Food price affects substitutions from consuming unhealthy to healthy food (Alwang et al., 2017). Besides, (Abdulai et al., 1999) mentioned that there are some factors that can have significant impact on food consumption in India such as region, urbanisation, household size, and seasonality. On the other hand,Gao et al. (2019)investigates that food consumption is affected by food availability, accessibility and choice. Similarly, the rise of income increases food consump-tion (Regmi et al., 2001), and food consumption has a significant impact on the water requirements, individual preferences, beliefs and cultural traditions (Liu and Savenije, 2008; Vasileskaa and Rechkoskaa, 2012). Moreover, women’s schooling and employ-ment have an important effect on the share of food in total
expen-diture and, in turn, an increase in food consumption (Abdulai and Aubert, 2004). Furthermore, various authors stated that climate change has an important impact on food consumption worldwide and food consumption patterns have changed with changes in food habits (Rodríguez et al., 2007; Popp et al., 2010; Hasegawa et al., 2013). Gerbens-Leenes et al. (2010) confirmed that for low-income countries, GDP increase is accompanied by changes in food consumption patterns, with a large gap between supply and actual consumption. Taking the mass media view,Musaiger (1993) con-firmed that televised food advertisements play a vital role in mod-ifying food habits.
Regarding seasonality, some authors address a specific season-ality of income and consumption (Khandker, 2012), and the sea-sonal fluctuation in consumption behaviour is conditioned by the presence of an off-farm source of income and high variability in consumption over the seasons (Handa and Mlay, 2006; Dercon and Krishnan, 2000). In most cases, the prices for seasonal food products fall at demand peaks (MacDonald, 2000). In the case of Africa, seasonality in food markets remains sizeable and the sea-sonal gap is highest among vegetables, fruits, and maize, and low-est among eggs and rice (Gilbert et al., 2017).
Substantial numbers of papers have investigated food price elasticities and examined various related demand models.Colen et al. (2018), using the meta-analysis of food elasticities combined with calorie and nutrients, estimated in previous studies in African countries, found that the estimates of income elasticities tend to be higher for studies using household expenditures as proxy for income and the size of the income elasticity becomes smaller as countries become richer. Whereas, (Shashika et al., 2019) studied the food price elasticities in Sri Lanka and found that the food own-price elasticities are negative and price inelastic. In addition, Colen et al. (2018)stated that food prices are the principal determi-nant of consumption patterns and usually food demand is income inelastic. Furthermore, a study conducted in Mali, using single-stage OLS regression, showed that the expenditure elasticities are significant and positive for the food commodities (Rogers and Lowdermilk, 1991). However, the removal of subsidies will lead to an increase in the expenditures on free market goods for all income groups. An additional study conducted in Ethiopia found that all major food grains have close to unitary own-price elastic-ities, and grains seem to be more price elastic for urban residents versus rural residents. Sorghum has been classified as an inferior food item among urban residents (Berhane et al., 2011). The elas-ticities for rice and for millet/sorghum are almost equal, indicating once again that one is not apparently inferior to the other (Rogers and Lowdermilk, 1991). An older, related study conducted by (Abdulai et al., 1999) used a linear approximate almost ideal demand system indicating that milk and milk products are income elastic in both rural and urban areas in India. In addition, meat, fish and eggs are income elastic in urban areas because of their rela-tively high share of the food income compared to rural areas and all other food groups that are income inelastic in India. Andreyeva et al. (2010)found that price elasticity for meat is pos-itive and inelastic. Actually, the urban regions are richer than rural regions, hence food elasticities are lower in urban areas compared to rural areas (Handa and Mlay, 2006).
Tan et al. (2017)used the QUAIDS to assess the distribution of living expenses on goods and services and showed that food expenditures have significantly affected other living expenses. Abdulai (2002)used QUAIDS and found that the own-price elastic-ity of fruits and vegetables, other foods and non-food are greater than unity, while the elasticity for the bread and cereals, meat and fish, milk, cheese, and eggs were price inelastic. Furthermore, Obayelu et al. (2009)performed the nonlinear QUAIDS model to estimate price and expenditure elasticities of food items consumed in Nigeria. Their outcomes showed that the own-price elasticities
Fig. 1. Food consumption in Africa (1990–2018). Sources: FAO (2019b), (FAO, 2019c)
of cereal crops, fruits, vegetables, and oil are price inelastic. Fur-thermore, own-price elasticities of most food groups in Tanzania are close to one, indicating a high response of quantities demanded to changes in food prices (Abdulai and Aubert, 2004). However, Hosni and Ramadan (2018)indicated that the cross-price elastici-ties for sugar and oil are almost zero between the free market and its rationed counterparts. Additionally, Lasarte et al. (2014) performed a study considering the consumption patterns of house-holds along different city sizes in the context of a developed coun-try using the AIDS. The authors confirmed that the size of the city in which the household resides has a similar significant and rele-vant effect on consumption patterns as family income level. Fur-thermore, the study confirmed that the consumption patterns are influenced by household place of residence, especially in urban or rural areas.
3. Data and methodology
The data used in this study was collected through a household-level survey during the period of 2017–2018 in a rural part of the eastern Nile River in Sudan using a multi-stratified sampling tech-nique through comprehensives questionnaires, which was dis-tributed randomly among households in the selected region. Table 1shows descriptive statistics of the questionnaire. All factor variables such as gender or marital status are coded one if the responded is men or married, respectively. Consequently, all dummy variables inTable 1have mean less than unity.
The eastern Nile River region was selected as the study area because the eastern Nile region comprises of various settled tribes including many internal forced displaced families and nomadic movements resulted from pastures conflicts. The displaced families migrated from drought areas (from north region) or form conflict area and civils wars areas (south, east and south regions) such as Jebel Marr, Blue Nile and south Kordofan, and about 1.86 million live in protracted displacement (UN, 2019) and settled around Khartoum, the open. Those households are more vulnerable to pov-erty and food insecurity.Abdelmoneium (2016)addressed various factors contributing to the internal displacement of housholds, which made them to migrate to Khartoum and other cities in Sudan. The conflict in Sudan has been driven by political and socio-cultural marginalisation, land dispossession, inter-ethnic conflict, repeated civil war, natural disasters such as famine, drought and floods in the north Sudan.
The data used in this study were obtained through question-naires administered to a random sample of 1000 households in the eastern Nile area. The targeted household live in such areas
are employed as farmers and majority of them migrated from the agricultural and pastoral lands.
The interview method is achieved by skilled numerators from Agricultural Economics and Policy Research Centre under profes-sional supervision and principal author. The sample included ran-domly selected households from all regions of the rural area in Eastern Nile. The data were collected at the end of the winter sea-son, i.e starting March 2018. The Methods to assess ‘‘usual” intake should rely on memory methods or methods that minimize changes to usual eating habit.Pennington J.A.T. (1991)identified 8 methods for obtaining food consumption information and included two folds: memory and non-memory methods. The study follows the memory methods that include frequencies and food consumption recall. Because it is difficult to follow the Non-memory methods to include food records, weighed intakes, and duplicate portions, the majority of the rural people unaware of the records methods. Moreover, the household food data expressed as food commodities consumed per household per week.
The questionnaires comprised social and economic items. The social items involved the socio-demographical status (age, educa-tion, family size, jobs, marital status, etc.) of each household head, while the economic items included the main food consumed by the households and food intake in terms of total consumption of the household kilocalories during three common seasons in Sudan, specifically winter, summer and autumn. Also, the survey included income (expenditures) and food prices. Due to political crisis and instability, the prices of the commodities vary from seller to other and out of control. After prices liberalization policy, the prices of consumer goods were out of control in Sudan and the dissimilarity of the commodities prices are common and widespread among regions. In general, the food prices is caused by conflict events in Sudan, particular cereals prices and wheat prices (Chen et al., 2018).
Descriptive analysis showed that sorghum food reports the highest percentage of consumption among households in Sudan (96% consume it in the summer, 73% consume it both autumn and winter seasons). 67% of the households consume wheat in the summer, which is the highest percentage compared to other seasons. Rice consumption was low and reached 7% of surveyed households. This finding reflects that rice is a luxury food, but con-sumed by both rich and poor people, and its consumption increases in urban areas. The surveyed households confirmed that rice consumption increase is associated with social events. This outcome is confirmed by a study performed in Mali (Rogers and Lowdermilk, 1991). The authors stated that rice is a luxury food consumed by wealthy people, while coarse grains (sorghum and millet) are largely consumed by poor people.
Table 1
Descriptive statistics.
Variable Obs Mean Std. Dev. Min Max
Gende (Male) 2,942 0.641 0.48 0 1
HH_Age 2,942 50 15 20 100
Marital status (married) 2,942 0.853 0.354 0 1
HH-size 2,942 7 3 1 18 Farmer 2,942 0.692 0.462 0 1 Housewife 2,942 0.189 0.391 0 1 Shepherd 2,942 0.022 0.148 0 1 Labour 2,942 0.015 0.12 0 1 Farmer&Shepherd 2,942 0.005 0.071 0 1 Driver 2,942 0.006 0.078 0 1 Trader 2,942 0.049 0.217 0 1 Teacher 2,942 0.017 0.131 0 1 PrimarySch 2,942 0.061 0.238 0 1 SecondarySch 2,942 0.04 0.197 0 1 HighSch 2,942 0.019 0.138 0 1 Bachelors 2,942 0.013 0.111 0 1 HH_Income 2,942 413 262 598 1659
Also, study results reveal that the sorghum consumption pat-terns depend on consumer preferences. Thus, it can be substituted by wheat and rice in urban areas, but in the rural areas, it is diffi-cult to substitute with other cereals. Additionally, we note that food consumption and preference is affected by households’ tribes in the surveyed sample. For instance, 80% of millet consumption comes from tribes that originated from the western regions, while the remaining (20%) comes from tribes that originated from other regions of Sudan. Regarding consumption of animal products, most of the surveyed households consumed meat in the summer (93%). This is true because meat prices start to decline in summer and autumn in Sudan due to availability of the animal fodders. In addi-tion, it is observed that around half of the households consume milk over the course of the year while less than half of the house-holds consume eggs, approximately 35%, in winter and autumn. These percentages increase in the summer to reach 63% of the respondents. Likewise, 34% of respondents consume chicken. Moreover, the households indicate that the decline of the animal products consumption is related to seasonal variations of the food prices and income.
As the study aims to analyse food elasticities, the appropriate model is QUAIDS (Banks et al., 1997), which is recommend by many authors (Denton and Mountain, 2004; Ecker and Qaim, 2008; Fashogbon and Oni, 2013; Guerrero-López et al., 2017) for analysing food demand. Moreover, QUAIDS has been implemented in African and Asian developing economies by estimating complete food demand systems (Sowunmi et al., 2020; Pallegedara, 2019). The empirical model of QUAIDS in the present study is derived from (Banks et al., 1997) as follows:
QUAIDS is derived basically from the indirect utility function (V) of the consumer given by:
In
v
ðy; pÞ ¼ Iny InaðpÞ bðpÞ 1 þ kðpÞ ( ) ð1Þ whereas y, denoted to be the total food expenditures for each house-hold, p is a vector of the nominal prices (p1---- p2), a(p) is a function that is homogenous of degree one in prices, and b(p) and k (p) are the distinct price aggregator functions that are homogenous of degree zero in prices. Likewise, ln a(p) and ln b(p) are stated as translog and cob-Douglass formulas as detailed in Deaton and Muellbauer’s AIDS model (1980). Furthermore, k (p) is set to zero in AIDS model (Deaton and Muellbauer, 1980).Thus, Ina pð Þ ¼
a
0þ Xn i¼1a
iInpiþ 1 2 Xn i¼1 Xn j¼1c
ijInpiInpj ð2Þ b pð Þ ¼Y n i pbi i ð3Þ kð Þ ¼p X n i¼1 kiInpi ð4ÞThe b(p) and k (p) are homogenous of degree zero in prices; then
v
(y, p) is homogenous of degree zero in both p and y. Equally, the equations theoretical parametric restrictions are as follows: Xn i¼1a
i¼ 1ðaddingupÞ ð5Þ Yn i pbi i ¼ 0; ðaddingupÞ ð6Þ Xn i¼1 ki¼ 0; ðaddingupÞ ð7Þ Xn i¼1c
ij¼ 0; ðhomogeneityÞ ð8Þ Xn j¼1c
ij¼ 0;c
ij¼c
ji; ðSlutskysymmetryÞ: ð9ÞThough, the theoretical restrictions of adding-up is fulfilled if the summation of the logarithmic form of Roy’s identity (wi) = 1, for all y and p via the application of the logarithmic form parame-ter. The Roy’s identity (wi):
wi¼ @ InV Inpi = @In
v
Iny ð10Þ Therefore, the QUAIDS resulting in the income shares form as: wi¼ @ InV Inpi = @InInyv
ð11Þ wi¼a
iþ Xn j¼1c
ijInpiþ biIn y a pð Þ þ ki b pð Þ In y bðpÞ 2 þe
i ð12Þwhereas wiexpresses the expenditure (budget share) for food i for each household, and the variables
a
i,c
ijand biare parameters to be estimated in the model, wherease
idenotes the error term.Earlier and recent articles (Tefera, 2012; Bairagi et al., 2020) argued that the demand for food consumption depends not only on the household income and food prices, but on other factors, which act as individual or mixed factors. These factors can be ter-med as qualitative factors (household preferences and socio-economics characters). In the current study, the households’ demo-graphical factors are incorporated in the QUAIDS model.
The study estimated Eq.(12)for six commodities demand sys-tem. As stated in the relevant literature (Abdulai and Aubert, 2004, Abdulai, 2002), ifki = 0 for all i, namely the quadratic term in all expenditure share equations drops out, then the model will be exactly the standard AIDS model proposed by Deaton and Muellbauer (1980).
The analysis incorporates demographic variables by using the scaling technique introduced byRay (1983)and extended to the quadratic AIDS model byPoi (2002). The study modifies the influ-ences of the household cost function so that prices and total expen-diture are scaled to reflect heterogeneity in household demographics (Poi, 2002), which generate the new adding-up con-dition. Then, for each household spending function He(p, x, u), according to Poi (2002), and subject to the budget shares, the expenditure function of the households’ reference Her (p, u) is scaled for the household demographic factors by the following equation:
y0ðp; x; uÞ ¼ y
z
ð Þ
u
ðp; x; uÞ ð13Þwhereas x, denotes the vector of the z characteristic and u is direct utility. The budget share Eq.(12)is specified with demographical effect: wi¼
a
iþ Xn j¼1c
ijInpjþ big
^A ixÞIn y y 0ðxÞa
ðpÞ " # þ ki bðpÞcðp; xÞ In y y0ðxÞa
ðpÞa
ðpÞ " # ( )2 ð14Þ whereas c(p, x) =Qnip g ^A jx i ,g
^A j, express the j thcolumn of the matrix
g
. The new adding-up condition necessitates thatPg
^Asj¼ 0:Identical to (Banks et al., 1997), and following (Cupák et al., 2015) formula, the expenditures and the price elasticities are
attained by partially differentiating the share budget equation with demographic effect (Eq. (14)), Iny and Inpj. Thus, the elasticities equations are computed as follows:
l
i¼ @x
i @Iny¼ biþg
^A ixþ 2ki bðpÞcðp; xÞIn y y 0ðxÞa
ðpÞ " # ð15Þl
ij¼ @x
i @Inpi ¼ yijl
ia
jþ X kc
jkInpk ! kiðbiþg
^A ixÞ bðpÞcðp; xÞ In y y 0ðxÞa
ðpÞ " # ( )2 ð16ÞFrom Eqs.(15) and (16), the expenditure equation is estimated as:
He¼
l
iwiþ 1 ð17Þ
whereas the uncompensated price elasticity (Marshallian) are esti-mated below:
Hu ij¼
lij
wi dij, dijexpresses Kronecker delta that equals one if i = j
and zero otherwise.
By applying the Slutsky equation, the compensated price elas-ticities (Hicksian) can be obtained as:
Hc
ij¼ H
u
ijþ Hewi ð18Þ
Table 2
Parameter estimates for the QUAIDS model for food consumption in rural Sudan. Variables Mean budget
shares
Constant Cereals Meat & chicken Milk & eggs Vegetables Staples Caffeine ıntake
Cereals 0.190 0.3411*** 0.1588*** (0.0075) (0.0027) Meat & chicken 0.614 0.1577*** (0.0115) 0.0953*** (0.0030) 0.2104*** (0.0048) Milk & eggs 0.085 0.2212***
(0.0055) 0.0165*** (0.0015) 0.0568*** (0.0022) 0.0758*** (0.0024) Vegetables 0.071 0.1604*** 0.0309*** 0.0325*** 0.0082*** 0.0652*** (0.0049) (0.0013) (0.0019) (0.0015) (0.0018) Staples 0.032 0.0815*** 0.0046*** 0.0191*** 0.0031*** 0.0000 0.0203*** (0.0025) (0.0007) (0.0009) (0.0009) (0.0008) (0.0009) Caffeine ıntake 0.008 0.0380 (0.3411) 0.0116*** (0.0006) 0.0066*** (0.0008) 0.0026*** (0.0007) 0.0064*** (0.0006) 0.0004 (0.0004) 0.0088*** (0.0004)
Variables Expenditure ExpenditureSQ HH-Size Gender HH-Age Marital Status PrimarySch Secondary Sch Cereals 0.0593*** 0.0272*** 0.0025*** 0.0053 0.0003 0.0172*** 0.0423*** 0.0504*** (0.0241) (0.0051) (0.0008) (0.0073) (0.0002) (0.0073) (0.0112) (0.0124) Meat& chicken 0.0082 (0.0353) 0.0669*** (0.0054) 0.0023* (0.0012) 0.0077 (0.0114) 0.0006*** (0.0002) 0.0131 (0.0109) 0.0547*** (0.0169) 0.0474*** (0.0191) Milk& eggs 0.0159 (0.0136) 0.0295*** (0.0031) 0.0024*** (0.0005) 0.0069 (0.0041) 0.0000 (0.0001) 0.0053 (0.0041) 0.0037 (0.0059) 0.0123** (0.0072) Vegetables 0.0607*** 0.0052* 0.0016*** 0.0004 0.0001 0.0066* 0.0034 0.0080 (0.0109) (0.0027) (0.0004) (0.0032) (0.0001) (0.0035) (0.0048) (0.0057) Staples 0.0034 0.0140*** 0.0010*** 0.0030 0.0000 0.0074*** 0.0146*** 0.0010 (0.0059) (0.0013) (0.0002) (0.0019) (0.0000) (0.0017) (0.0031) (0.0034) Caffeine ıntake 0.0097** (0.0048) 0.0090*** (0.0014) 0.0001 (0.0002) 0.0027** (0.0013) 0.0001 (0.0000) 0.0004 (0.0014) 0.0024 (0.0020) 0.0004 (0.0024)
Variables HighSch Bachelors South East West Farmer Housewife Shepherd Cereals 0.0286 0.0653*** 0.0776*** 0.0745*** 0.0520*** 0.0802*** 0.1087*** 0.0922*** (0.0207) (0.0163) (0.0084) (0.0085) (0.0091) (0.0191) (0.0207) (0.0246) Meat& chicken 0.0279 (0.0300) 0.1114*** (0.0260) 0.1684*** (0.0123) 0.1359*** (0.0125) 0.1204*** (0.0143) 0.1351*** (0.0296) 0.1484*** (0.0318) 0.1534*** (0.0369) Milk& eggs 0.0317*** (0.0104) 0.0481*** (0.0100) 0.0704*** (0.0046) 0.0355*** (0.0046) 0.0202*** (0.0050) 0.0507*** (0.0112) 0.0477*** (0.0120) 0.0312*** (0.0140) Vegetables 0.0029 0.0036 0.0171*** 0.0171*** 0.0370*** 0.0108*** 0.0172* 0.0203* (0.0094) (0.0073) (0.0040) (0.0041) (0.0041) (0.0084) (0.0091) (0.0113) Staples 0.0368*** 0.0172*** 0.0045** 0.0116*** 0.0164*** 0.0191*** 0.0159*** 0.0594*** (0.0064) (0.0050) (0.0021) (0.0021) (0.0024) (0.0049) (0.0052) (0.0065) Caffeine ıntake 0.0029 (0.0040) 0.0229*** (0.0032) 0.0012 (0.0017) 0.0028 (0.0017) 0.0053*** (0.0017) 0.0040 (0.0037) 0.0068 (0.0040) 0.0091* (0.0049) Variables Labour Farmer&shepherd Driver Trader Teacher Cereals 0.0672*** 0.0864*** 0.0117 0.1207*** 0.1631*** (0.0246) (0.0361) (0.0375) (0.0237) (0.0238) Meat&chicken 0.1270*** 0.2119*** 0.0651 0.1596*** 0.2112*** (0.0398) (0.0537) (0.0499) (0.0357) (0.0368) Milk&eggs 0.0667*** 0.0485*** 0.0965*** 0.0328*** 0.0423*** (0.0152) (0.0201) (0.0180) (0.0133) (0.0137) Vegetables 0.0301*** 0.0145 0.0335** 0.0145 0.0105 (0.0106) (0.0168) (0.0155) (0.0105) (0.0097) Staples 0.0370*** 0.1015*** 0.0098 0.0223*** 0.0254*** (0.0074) (0.0108) (0.0092) (0.0060) (0.0061) Caffeine ıntake 0.0138*** 0.0100 0.0194*** 0.0017 0.0091** (0.0048) (0.0072) (0.0074) (0.0046) (0.0045) Note: Standard errors are in parentheses. ***, **, * denote significance at 1%, 5%, 10% levels, respectively.
4. Results and discussion
The demand system consisted of six expenditure groups: cere-als (sorghum, millet, wheat), meat and chicken, milk and eggs, veg-etables (okra, onion, tomato), staples (sugar, salt, oil), and caffeine intake (tea and coffee). The demographic variables included are household size, age, gender, educational attainment (no schooling, primary school, secondary school, high school, bachelors), occupa-tion of household head (farmer, housewife, shepherd, labour, farmer and shepherd, driver, trader, teacher, others), and geo-graphical locations (northern, southern, eastern, western regions). One of the subgroups for each demographic variable in categorical form is excluded in the estimation. People having no schooling, occupation of household head described as ‘‘others,” and northern region are excluded to avoid singularity.
The six equations demand system inTable 2are estimated by iterated feasible generalized nonlinear least-squares estimation. The parameter estimates are invariant to the equation dropped from the demand system estimation. The analysis incorporates demographic factors by using the scaling technique introduced byRay (1983)and extended to the quadratic AIDS model byPoi (2002).
The theoretical restrictions, namely, adding up, homogeneity, and symmetry are maintained in the estimation. The coefficients of quadratic term (ExpenditureSQ) are statistically significant at 5% level or better for all demand equations inTable 2. Therefore, it states a clear evidence that QUAIDS model is preferred over stan-dard AIDS model estimation. The results show that men spend more from their income on caffeine compared to women. Married households spend less from their income on cereal and staples compared to unmarried households. Furthermore, respondents whose education level is primary school spend more from their income on meat and chicken and spend less on cereals and staples compared to illiterate respondents. Moreover, secondary school educated households spend less from their income on meat & chicken, milk & eggs and spend more on cereals compared to illit-erate households. Also, both high school diploma and bachelor’s degree holders spend more from their income on milk and eggs compared to the illiterate. Also, the results show that place of res-idence has an impact on household budget share. For example, household who reside in the south and the east of Sudan spend more on meat and chicken compared to households who live in the north of Sudan.
The model estimation results inTable 2indicate that the quad-ratic terms for all equations of the demand system are statistically significant at 5% level or better. In addition, we test H0 :ki = 0 for all i which allows us to select the preferred model between the
AIDS and the QUAIDS models. The test results confirm that QUAIDS is more suitable for Sudanese food demand data.
(X2(5) = 120.30, Prob > chi2 = 0.0001).
Table 3reports the Marshallian price elasticities and expendi-ture elasticities. All estimated own-price elasticities are statisti-cally significant at 5% level or better and have theoretistatisti-cally consistent signs as the own-price elasticities are negative, except the caffeine intake equation. Also, the own-price elasticities are inelastic indicating that changes in prices have small impact on quantity demanded. In addition, expenditure elasticities are posi-tive and statistically significant at 5% level or better. The expendi-ture elasticity for meats and chicken is elastic (1.34) suggesting that meats and chicken may be classified as a luxury good in Sudan. This means that, on average, a one percent increase in rural Sudanese household income increases chicken consumption by 1.34 percent. This result is consistent with (Ibrahim, 2018) find-ings, which showed that elasticity of meat demand in Sudan is income elastic.
Remaining expenditure elasticities are less than 1, hence, all other food groups are necessities. Furthermore, the cross-price elasticities show that most items have complementary relation-ship in consumption.
5. Conclusions
This paper uses the QUAIDS model to estimate Sudanese house-holds food consumption by focusing on six food items namely; cereals (sorghum, millet, wheat), meat and chicken, milk and eggs, vegetables (okra, onion, tomato), staples (sugar, salt, oil), and caf-feine intake (tea and coffee). The results of demographical vari-ables show that men spend more from their income on caffeine (tea and coffee) compared to women. Also, married respondents spend less on cereals and staples compare to unmarried respon-dents. Also, the results reveal that place of residence indeed has an impact on household consumption. For instance, households who live in the south and east of Sudan spend more on meat and chicken compared to households who live in the north, while households who live in the west spend less from their income on meat and chicken compared to households who live in the north of Sudan. Also, the estimated Marshallian own-price elasticities of demand were all price inelastic for all commodities, indicating that price changes have marginal impact on quantity demanded. Also, income elasticities show that all food items were necessities, except meat and chicken, which are classified as luxury items. Also, the cross-price elasticities show that nearly a majority of the food items are complementary in consumption by Sudanese households.
Table 3
Marshallian price elasticities and expenditure elasticities estimates for the QUAIDS model for food consumption in rural Sudan.
Variables Cereals Meat& chicken Milk& eggs Vegetables Staples Caffeine ıntake Expenditure elasticities Cereals 0.0695*** 0.0713*** 0.0435*** 0.1257*** 0.0103*** 0.0563*** 0.3766*** (0.0143) (0.0196) (0.0080) (0.0071) (0.0037) (0.0032) (0.0149) Meat& chicken 0.2114*** (0.0048) 0.9153*** (0.0088) 0.1186*** (0.0035) 0.0752*** (0.0031) 0.0396*** (0.0015) 0.0136*** (0.0013) 1.3737*** (0.0060) Milk& eggs 0.1116*** (0.0174) 0.3045*** (0.0281) 0.0770*** (0.0287) 0.0641*** (0.0179) 0.0481*** (0.0111) 0.0347*** (0.0085) 0.4745*** (0.0183) Vegetables 0.3529*** 0.0787*** 0.0779*** 0.0520** 0.0118 0.0934*** 0.4562*** (0.0187) (0.0300) (0.0214) (0.0249) (0.0117) (0.0089) (0.0202) Staples 0.0292 0.0812*** 0.1484*** 0.0448* 0.3502*** 0.0180 0.2494*** (0.0215) (0.0327) (0.0296) (0.0261) (0.0293) (0.0129) (0.0206) Caffeine ıntake 1.4061*** (0.0756) 0.5286*** (0.1184) 0.3555*** (0.0918) 0.8236*** (0.0808) 0.0570 (0.0524) 0.1108** (0.0538) 0.5878*** (0.0806)
Declaration of Competing Interest
The authors declare that they have no known competing finan-cial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgement
The authors would like to thank Victoria J. Tice for English edit-ing and proofreadedit-ing.
References
Abdelmoneium, A., 2016. Internally displaced families in Khartoum-Sudan: challenges and policy implications. Int. J. Migration, Health and Social Care 12
(4), 252–263. https://doi-org.sdl.idm.oclc.org/10.1108/IJMHSC-05-2015-0017.
Abdulai, A., 2002. Household demand for food in Switzerland. A quadratic almost ideal demand system. Swiss J. Econ. Statistics (SJES) 138 (I), 1–18.
Abdulai, A., Aubert, D., 2004. A cross-section analysis of household demand for food and nutrients in Tanzania. Agricultural Econ. 31 (1), 67–79. https://doi-org.sdl.
idm.oclc.org/10.1111/j.1574-0862.2004.tb00222.x.
Abdulai, A., Jain, D.K., Sharma, A.K., 1999. Household food demand analysis in India. J. Agric. Econ. 50 (2), 316–327. https://doi.org/10.1111/j.1477-9552.1999. tb00816.x.
Ali Taha, S.A., 1978. Household food consumption in five villages in the Sudan. Ecol. Food Nutr. 7 (3), 137–142.https://doi.org/10.1080/03670244.1978.9990523. Alwang, J., Gotor, E., Thiele, G., Hareau, G., Jaleta, M., Chamberlin, J., 2017. Pathways
from research on improved staple crop germplasm to poverty reduction for smallholder farmers. Agric. Syst. 172 (1), 16–27. https://doi.org/10.1016/j. agsy.2017.10.005.
Andreyeva, T., Long, M.W., Brownell, K.D., 2010. The impact of food prices on consumption: a systematic review of research on the price elasticity of demand for food. Am. J. Public Health 100 (2), 216–222. https://doi.org/10.2105/
AJPH.2008.151415.
Banks, J., Blundell, R., Lewbel, A., 1997. Quadratic Engel curves and consumer demand. Rev. Econ. Stat. 79 (4), 527–539. https://doi.org/10.1162/
003465397557015.
Bairagi, S., Mohanty, S., Baruah, S., Thi, H.T., 2020. Changing food consumption patterns in rural and urban Vietnam: Implications for a future food supply system. Australian J. Agricultural Resour. Econ. https://doi-org.sdl.idm.oclc.org/
10.1111/1467-8489.12363.
Berhane, G., Paulos, Z., Tafere, K., Tamru, S., 2011. Foodgrain consumption and calorie intake patterns in Ethiopia. IFPRI Ethiopia Strategy Support Program II (ESSP II). Working Paper 23.
Chen, J., Kibriya, S., Bessler, D., Price, E., 2018. The relationship between conflict events and commodity prices in Sudan. J. Policy Modeling 40(4), 663-684. https://doi-org.sdl.idm.oclc.org/10.1016/j.jpolmod.2018.01.014
Cupák, A., Pokrivcˇák, J., Rizov, M., 2015. Food demand and consumption patterns in the new eu member states: the case of Slovakia. Ekonomicky´ cˇasopis (J. Econ.)
63 (4). http://dx.doi.org/10.22004/ag.econ.182825.
Colen, L., Melo, P., Abdul-Salam, Y., Roberts, D., Mary, S., Paloma, S.G.Y., 2018. Income elasticities for food, calories and nutrients across Africa: A meta-analysis. Food Policy 77, 116–132. https://doi.org/10.1016/ j.foodpol.2018.04.002.
Delgado, C.L., 2003. Rising consumption of meat and milk in developing countries has created a new food revolution. J. Nutr. 133 (11), 3907S–3910S.https://doi.
org/10.1093/jn/133.11.3907S.
Deaton, S., Muellbauer, J., 1980. An almost ideal demand system. Am. Econ. Rev. 70 (3), 312–326https://www.jstor.org/stable/1805222.
Denton, F.T., Mountain, D.C., 2004. Aggregation effects on price and expenditure elasticities in a quadratic almost ideal demand system. Canadian Journal of Economics/Revue canadienne d’économique 37 (3), 613–628.https://doi.org/
10.1111/j.0008-4085.2004.00240.x.
Dercon, S., Krishnan, P., 2000. Vulnerability, seasonality and poverty in Ethiopia. J. Dev. Stud. 36 (6), 25–53.https://doi.org/10.1080/00220380008422653. Ecker, O., Qaim, M., 2008. Income and Price Elasticities of Food Demand and
Nutrient Consumption in Malawi. Annual Meeting, July 27-29, 2008, Orlando, Florida (No. 6349). Agricultural and Applied Economics Association). Retrieved from. https://ideas.repec.org/p/ags/aaea08/6349.html
FAO., 2009. The 2007–2008 food price episode Impact and policies in Eastern and Southern Africa. Food and Agriculture Organization of the United Nations. http://www.fao.org/3/a-i0984e.pdf
FAO, 2019a. Country Programming Framework for the Republic of Sudan CPF (2012–2016). http://www.fao.org/sudan/programmes-and-projects/en/ FAO, 2019b. AQUASTAT - FAO’s Global Information System on Water and
Agriculture. http://www.fao.org/nr/water/aquastat/data/query/results.html FAO, 2019c. Food Composition data. http://www.fao.org/faostat/en/#data FAO-SIFSIA, 2010. Sudan Institutional Programme, FAO/SIFSIA Programme in
northern Sudan. Food Security Information for Action (SIFSIA). Rome. http:// www.fao.org/3/a-i4388e.pdf
Fashogbon, A.E., Oni, O.A., 2013. Heterogeneity in rural household food demand and its determinants in Ondo State, Nigeria: An application of quadratic almost ideal demand system. J. Agric. Sci. 5 (2), 169http://ccsenet.org/journal/index.php/jas/
article/view/21696/15204.
Ibrahim, A.M., 2018. Food behavior and consumption pattern in rural white Nile of Sudan. Int. J. Agric. Sc. Food Technol. 4 (1), 007–014.https://doi.org/10.17352/
2455-815X.000030.
Gao, B., Huang, W., Wang, L., Huang, Y., Ding, S., Cui, S., 2019. Driving forces of nitrogen flows and nitrogen use efficiency of food systems in seven Chinese cities, 1990 to 2015. Sci. Total Environ. 676, 144–154.https://doi.org/10.1016/j. scitotenv.2019.04.136.
Gerbens-Leenes, P., Nonhebel, S., Krol, M.S., 2010. Food consumption patterns and economic growth. Increasing affluence and the use of natural resources. Appetite 55 (3), 597–608.https://doi.org/10.1016/j.appet.2010.09.013. Gilbert, C.L., Christiaensen, L., Kaminski, J., 2017. Food price seasonality in Africa:
Measurement and extent. Food Policy 67, 119–132.https://doi.org/10.1016/ j.foodpol.2016.09.016.
Green, R., Cornelsen, L., Dangour, A.D., Turner, R., Shankar, B., Mazzocchi, M., Smith, R.D., 2013. The effect of rising food prices on food consumption: systematic review with meta-regression. BMJ 346,. https://doi.org/10.1136/bmj.f3703
f3703.
Guerrero-López, C.M., Unar-Munguía, M., Colchero, M.A., 2017. Price elasticity of the demand for soft drinks, other sugar-sweetened beverages and energy dense food in Chile. BMC Public Health 17 (1), 180. https://doi.org/10.1186/s12889-017-4098-x.
Handa, S., Mlay, G., 2006. Food consumption patterns, seasonality and market access in Mozambique. Development Southern Africa 23 (4), 541–560.https://
doi.org/10.1080/03768350600929249.
Hasegawa, T., Fujimori, S., Shin, Y., Takahashi, K., Masui, T., Tanaka, A., 2013. Climate change impact and adaptation assessment on food consumption utilizing a new scenario framework. Environ. Sci. Technol. 48 (1), 438–445.https://doi.org/
10.1021/es4034149.
Hassan, R.M., Babu, S.C., 1991. Measurement and determinants of rural poverty: Household consumption patterns and food poverty in rural Sudan. Food Policy 16 (6), 451–460.https://doi.org/10.1016/0306-9192(91)90045-L.
Hosni, R., Ramadan, R., 2018. Food subsidy or cash transfer: Impact of the food subsidy reform on Egyptian households. New Medit: Mediterranean Journal of Economics, Agriculture and Environment. Revue Méditerranéenne D’economie, Agriculture et Environment 17 (3), 9–22.
Humphries, D.L., Dearden, K.A., Crookston, B.T., Woldehanna, T., Penny, M.E., Behrman, J.R., 2017. Household food group expenditure patterns are associated with child anthropometry at ages 5, 8 and 12 years in Ethiopia, India, Peru and Vietnam. Econ. Hum. Biol. 26, 30–41. https://doi.org/10.1016/j. ehb.2017.02.001.
Ibnouf, F.O., 2009. The role of women in providing and improving household food security in Sudan: Implications for reducing hunger and malnutrition. J. Int. Women’s Stud. 10 (4), 144–167https://vc.bridgew.edu/jiws/vol10/iss4/ 10/.
Ibnouf, F.O., 2011. Challenges and possibilities for achieving household food security in the Western Sudan region: the role of female farmers. Food Security 3 (2), 215–231 https://link.springer.com/article/10.1007/s12571-011-0118-3.
Kearney, J., 2010. Food consumption trends and drivers. Philos. Trans. Royal Soc. B: Biol. Sci. 365 (1554), 2793–2807.https://doi.org/10.1098/rstb.2010.0149. Khalid, F.A., Ali, A.K., Ali, S.A., Mosmar, Z.Y., Salih, S.S., Salman, T.K., Mohammed, A.
A., 2017. Households’ dietary habits and food consumption patterns in Hamishkoreib locality, Kassala State, Sudan. J. Ethnic Foods 4 (3), 181–186.
https://doi.org/10.1016/j.jef.2017.08.009.
Khandker, S.R., 2012. Seasonality of income and poverty in Bangladesh. J. Dev. Econ. 97 (2), 244–256.https://doi.org/10.1016/j.jdeveco.2011.05.001.
Lasarte, E., Rubiera, F., Paredes, D., 2014. City size and household food consumption: demand elasticities in Spain. Appl. Econ. 46 (14), 1624–1641.https://doi.org/
10.1080/00036846.2013.868593.
Li, Y., Yu, W., 2010. Households food security in poverty-stricken regions: evidence from western rural China. Agric. Agric. Sci. Procedia 1, 386–395.https://doi.org/ 10.1016/j.aaspro.2010.09.048.
Liu, J., Savenije, H.H.G., 2008. Food consumption patterns and their effect on water requirement in China. Hydrol. Earth Syst. Sci. Discuss. 12 (3), 887–898https:// hal.archives-ouvertes.fr/hal-00305175.
MacDonald, J.M., 2000. Demand, information, and competition: why do food prices fall at seasonal demand peaks?. J. Ind. Econ. 48 (1), 27–45.https://doi.org/
10.1111/1467-6451.00111.
Mahgoub, F., 2014. Current status of agriculture and future challenges in Sudan: Nordiska Afrikainstitutet. Nordiska Afrikainstitutet, Uppsala, Sweden, p. 98. Musaiger, A.O., 1993. Socio-cultural and economic factors affecting food
consumption patterns in the Arab countries. J. Royal Soc. Health 113(2), 68-74. https://doi.org/10.1177%2F146642409311300205
Pennington, J.A.T., 1991. Methods for obtaining food consumption information. In: Macdonald, I. (Ed.), Monitoring Dietary Intakes. ILSI Monographs. Springer,
London. https://doi-org.sdl.idm.oclc.org/10.1007/978-1-4471-1828-2_1.
Obayelu, A., Okoruwa, V., Ajani, O., 2009. Cross-sectional analysis of food demand in the North Central, Nigeria: The quadratic almost ideal demand system (QUAIDS) approach. China Agricultural Econ. Rev. 1 (2), 173–193.https://doi.
org/10.1108/17561370910927426.
Pallegedara, A., 2019. Food consumption choice and demand by the Sri Lankan households: Trends, drivers and policy implications. J. Agribusiness Dev.
Emerging Econ. 9 (5), 520–535.
https://doi-org.sdl.idm.oclc.org/10.1108/JADEE-01-2019-0014.
Poi, B.P., 2002. From the help desk: Demand system estimation. Stata J. 2(4), 403– 410. https://doi.org/10.1177%2F1536867X0200200406.
Popp, A., Lotze-Campen, H., Bodirsky, B., 2010. Food consumption, diet shifts and associated non-CO2 greenhouse gases from agricultural production. Global Environ. Change 20 (3), 451–462. https://doi.org/10.1016/ j.gloenvcha.2010.02.001.
Ray, R., 1983. Measuring the costs of children: an alternative approach. J. Public Econ. 22 (1), 89–102.https://doi.org/10.1016/0047-2727(83)90058-0. Regmi, A., Deepak, M., Seale Jr, J. L., & Bernstein, J., 2001. Cross-country analysis of
food consumption patterns. Changing structure of global food consumption and trade, 1422. https://naldc.nal.usda.gov/download/34108/PDF#page=18 Rodríguez, C., Naves, J., Fernández-Gil, A., Obeso, J.R., Delibes, M., 2007. Long-term
trends in food habits of a relict brown bear population in northern Spain: the influence of climate and local factors. Environ. Conserv. 34 (1), 36–44.https://
doi.org/10.1017/S0376892906003535.
Rogers, B.L., Lowdermilk, M., 1991. Price policy and food consumption in urban Mali. Food Policy 16 (6), 461–473. https://doi-org.sdl.idm.oclc.org/10.1016/
0306-9192(91)90046-M.
Siddig, K., 2010. Macroeconomy and agriculture in Sudan: Analysis of trade policies, external shocks, and economic bans in a computable general equilibrium approach. Available at SSRN 1684084. https://ssrn.com/abstract=1684084 Shashika D., Rathnayaka, Selva Selvanathan A, E., Selvanathan, Saroja, 2019.
Consumption patterns in Sri Lanka: a decomposition analysis. Applied Economics 51, 4056–4072. https://doi.org/10.1080/00036846.2019.1588950. In this issue.
Sowunmi Fatai, A. et al., 2020. Household demand for basic foodstuffs in a recessed economy: a case study of Southwest Nigeria. World J. Sci., Technol. Sustain. Dev.
17 (1), 1–19. https://doi-org.sdl.idm.oclc.org/10.1108/WJSTSD-06-2019-0029.
Tan, T., Zhang, Y., Wen, B., Zhang, X., Zhan, J., 2017. Household consumption distribution in rural China: A consistent two-step estimation. Can. J. Agric. Économiques/Revue canadienne d’agroéconomie 65 (1), 119–133.https://doi.
org/10.1111/cjag.12103.
Tefera, N., 2012. Welfare impacts of rising food prices in rural Ethiopia: a Quadratic almost ideal demand system approach (No. 1007-2016-79682). https:// ageconsearch.umn.edu/record/124861
UN., 2019. United nation, Humanitarian Response Plan Report, Sudan. https:// reliefweb.int/sites/reliefweb.int/files/resources/
Sudan_2019_Humanitarian_Response_Plan.pdf
Vasileska, A., Rechkoska, G., 2012. Global and regional food consumption patterns and trends. Procedia-Social Behav. Sci. 44, 363–369.https://doi.org/10.1016/j. sbspro.2012.05.040.
Von Braun, J., Zaki, E., 1992. Drought and famine relationships in Sudan: policy implications. Food Nutr. Bull. 14 (2), 1–3https://www.cabdirect.org/cabdirect/
abstract/19916712073.
World Bank., 2019. Data Bank. World Development Indicators. https://data.worldbank.org/country/sudan
Zezza, A., Carletto, C., Fiedler, J.L., Gennari, P., Jolliffe, D., 2017. Food counts. Measuring food consumption and expenditures in household consumption and expenditure surveys (HCES). Introduction to the special issue. Food Policy 72, 1– 6.https://doi.org/10.1016/j.foodpol.2017.08.007.
Further reading
Rathnayaka, S.D., Selvanathan, E.S., Selvanathan, S., 2019a. Consumption patterns in Sri Lanka: a decomposition analysis. Appl. Econ. 1–17.https://doi.org/10.1080/
