The Journal of Development Studies
ISSN: 0022-0388 (Print) 1743-9140 (Online) Journal homepage: https://www.tandfonline.com/loi/fjds20
The Urban Food Question in the Context of
Inequality and Dietary Change: A Study of
Schoolchildren in Accra
Sara Stevano, Deborah Johnston & Emmanuel Codjoe
To cite this article: Sara Stevano, Deborah Johnston & Emmanuel Codjoe (2019): The Urban
Food Question in the Context of Inequality and Dietary Change: A Study of Schoolchildren in Accra,
The Journal of Development Studies, DOI: 10.1080/00220388.2019.1632434
To link to this article:  https://doi.org/10.1080/00220388.2019.1632434
Published online: 26 Jun 2019.
Submit your article to this journal 
Article views: 102
View related articles 
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=fjds20
The Journal of Development Studies, 2019
https://doi.org/10.1080/00220388.2019.1632434
The Urban Food Question in the Context of
Inequality and Dietary Change: A Study of
Schoolchildren in Accra
SARA STEVANO*, DEBORAH JOHNSTON ** & EMMANUEL CODJOE†
*Department of Accounting, Economics and Finance, University of the West of England (UWE), Bristol, UK, **Department
of Economics, SOAS University of London, London, UK, †Department of Economics, University of Ghana, Legon, Ghana
(Original version submitted July 2018; final version accepted June 2019)
ABSTRACT Diets are changing globally, as agricultural and food systems have become globalised.
Understanding how patterns of globalisation affect welfare is a key development question, but we know little
about the way that the globalisation of food systems impacts different groups. This study explores food security
and consumption among schoolchildren in Accra. We use a novel approach based on triangulation of primary
data on food consumption and a synthesis of secondary literature on food trade, policy and urban food
environment. Thus, we bridge a divide between micro-level analyses of food consumption and macro-level
studies of food systems. We find that socio-economic status is a critical dimension, with poorer children more
vulnerable to food insecurity and narrow dietary diversity. However, the consumption of packaged and processed
foods, often sugar-rich and nutrient-poor, cuts across wealth groups. We argue that the urban food question
today is defined by two intersecting phenomena: inequality and dietary change. The urban poor continue to face
the fundamental challenge of adequate food access amidst a food environment that provides consumers with
unhealthy and cheap food options. Therefore, food policy needs to regulate imports of cheap, unhealthy and
enticing food.
1. Introduction
Urban diets are changing, as agricultural and food systems have become globalised and created new
forms of food production, distribution, and trade (Hawkes, Blouin, Henson, Drager, & Dubé, 2010;
Lerner, Sweeney, & Eakin, 2014). The nutritional implications for the Global South are seen in rising
obesity and non-communicable diseases alongside persisting food insecurity and old-fashioned food
policies (Masters, Rosenblum, & Alemu, 2018; Popkin, Adair, & Ng, 2012). Understanding how
patterns of globalisation affect the welfare of populations is a key development question, but we
know little about the way that the globalisation of food and agriculture systems affects different
individuals or groups. Hawkes (2006) has suggested that dietary change will be experienced
differently by rich and poor, by urban and rural populations. However, more context-specific
evidence is needed to capture the complexity of impacts, and to frame the urban food question in
the economic, social and political factors that shape urban food systems.
This study explores food consumption among schoolchildren from different socio-economic back-
grounds in Accra, using a novel approach based on triangulation of primary quantitative and
qualitative data on food consumption and a synthesis of secondary literature on food trade, policy
Correspondence Address: Sara Stevano, Department of Accounting, Economics and Finance, University of the West of
England (UWE), Bristol, UK. Email: Sara.Stevano@uwe.ac.uk
© 2019 Informa UK Limited, trading as Taylor & Francis Group
2 S. Stevano et al.
and urban food environment. Thus, we bridge a divide between micro-level analyses of food
consumption and macro-level studies of food systems. This paper has two main objectives. First, it
seeks to extend the literature on urban food consumption through a case study of schoolchildren,
given the gap that persists in our knowledge on food access in the urban South (Floro & Swain, 2013;
Lagerkvist, Hess, Okello, & Karanja, 2013). Second, the paper interrogates the urban food question
today, in the context of rising inequality and global dietary change. Thus, it engages with studies on
inequality (Agyire-Tettey, Ackah, & Asuman, 2018; Kollamparambil, 2017) and on the nature of
urban food security in African cities in the 1990s (Levin et al., 1999; Maxwell, 1999; Maxwell et al.,
2000). Finally, this paper proposes that to better understand urban food consumption, we need an
integration of the research approaches used in public health and the political economy of food
provisioning.
We find that socio-economic status is a critical dimension of food consumption, with poorer
children more vulnerable to food insecurity and narrow dietary diversity. However, the consumption
of packaged and processed foods, often sugar-rich and nutrient-poor, cuts across wealth groups. This
illustrates the widespread desirability and availability of cheap, ultra-processed foods, with conco-
mitant implications for the future welfare of urban populations in developing countries (Monteiro,
Moubarac, Cannon, Ng, & Popkin, 2013). Thus, industrial and trade policies that regulate food
processing and content, as well as production and imports of cheap unhealthy foods are much-needed.
The next section provides the background on agricultural policy, urban food security and dietary
change in Ghana, highlighting how concerns about urban food security have changed over time. The
third section describes the study setting and the methodology. The fourth section presents the findings
on urban food provisioning (4.1), food security (4.2), food consumption (4.3) and food knowledge
(4.4). The final section discusses the results and concludes.
2. Agricultural policy, dietary change and urban food security in Ghana
Between 1970 and 1982, agricultural policy in Ghana had at its core Operation Feed Yourself (OFY),
aimed at increased production of maize and rice to achieve national food self-sufficiency, especially
for the growing urban population and the industrial sector (Girdner, Olorunsola, Froning, & Hansen,
1980; Nyanteng & Wayo Seini, 2000). Rice production increased and cocoa production, previously
the dominant export crop, declined (Nyanteng & Wayo Seini, 2000). Girdner et al. (1980) argue that,
while the agricultural policy of the 1970s promoted domestic food production, Ghana remained
overly reliant on colonial export commodities – cocoa and palm oil – and, by consequence, had to
pay a high bill for food imports.
The Ghanaian economy was hit by macroeconomic and political instability in the second half of
the 1970s, which resulted in urban protests driven by food shortages and high food prices (Maxwell,
1999) and led to the implementation of the Economic Recovery Programme (ERP) in 1983, a set of
reforms supported by the International Monetary Fund. At first, the ERP for the agricultural sector
sought to maintain self-sufficiency in cereals production but also to boost cocoa exports (Nyanteng &
Wayo Seini, 2000). In later phases of the programme, focus shifted to non-traditional export crops
due to the decline of traditional commodity prices (Amanor & Chichava, 2016; Ouma, 2015). With
the removal of agricultural subsidies and later expansion of export horticulture, food imports
continued to grow (Amanor & Chichava, 2016).
The structural adjustment also caused rising unemployment and declining urban wages, thus
affecting urban population’s ability to buy food (Maxwell, 1999). In Accra, households responded
by diversifying income sources and taking up various paid jobs and cash-earning activities in the
informal economy (Levin et al., 1999; Maxwell, 1999). Women’s increased participation in income-
generating activities, such as petty trading and food preparation for sale, and increasing prominence
of street food have been seen as outcomes of this period of economic liberalisation (Ibid.).
Thus, reliance on food imports and informalisation of employment, which has shaped food supply
in urban areas through, inter alia, the expansion of street food and imported (packaged) foods, are
The urban food question in Ghana 3
key features of the Ghanaian food environment from the 1990s to the present. Based on broad
national data, we know that Ghana produces approximately 50 per cent of its cereal and meat needs
and spends one billion USD on food imports (Amanor & Chichava, 2016). While the country is
estimated to be self-sufficient in maize, cassava and yam production, it is dependent on imports of
wheat and rice (Cudjoe, Breisinger, & Diao, 2010). Over 70 per cent of rice, now the major staple in
Ghana, is imported mostly from Thailand and Vietnam (Amanor, 2015). Cudjoe et al. (2010) find that
during the 2007–2008 food crisis, when international prices of wheat and rice sharply increased, the
urban poor were the hardest hit because urban food prices follow more closely the movements of
world prices. Thus, the food crisis has exposed the country’s vulnerability due to dependence on food
imports, and that the urban poor are at risk of food insecurity.
What are the implications for patterns of urban diets? Over the past few decades, progress in
reducing hunger and undernutrition has been remarkable in Ghana, with the Global Hunger Index
(GHI) decreasing from 42.7 in 1992 to 13.9 in 2016 (GHI, 2016).1 Regional data suggest that child
undernutrition is highest in the poorer regions of the North and lowest in Greater Accra
(Demographic and Health Survey [DHS], 2014). At the same time, overweight and obesity are
increasing (Ofori-Asenso, Agyeman, Laar, & Boateng, 2016). Overnutrition appears to affect pre-
dominantly urban and wealthier individuals, and it is higher among women than men (DHS, 2014).
However, national and regional averages do not illuminate the pathways of change and leave us with
no tools to identify the most vulnerable.
Dietary change is underway in Ghana. Broad brush estimates suggest that staple consumption is
changing in composition, with a greater prevalence of rice, which shows increasing trends in both
production and imports, especially in the early 2000s (FAOSTAT 2017). However, the share of dietary
energy supply derived from cereals, roots and tubers is decreasing (Ibid.), which indicates that diets are
becoming less reliant on cereals and more reliant on other foodstuff.2 This is in line with data on caloric
intake on developing countries as a group suggesting that relative importance of cereals is stagnant or
decreasing, and that of vegetable oils, sugar and meat is increasing (Food and Agriculture Organisation
[FAO], 2015). In addition, staple consumption is shifting towards rice, with increasing trends in rice
production and imports (FAOSTAT 2017). National-level data on food trade in Ghana also indicate that
imports of milk, dairy products, poultry and sugar are on an upward trend since the mid-1990s. Imports of
vegetable oils and fruit have exceeded exports since the early 2000s (FAOSTAT 2017). These patterns
may suggest that diets are diversifying; however, there are also concerns about the deterioration of diet
quality, with increased consumption of sugar-rich, nutrient-poor and ultra-processed foods, which is in
turn linked inter alia to globalisation and liberalisation of food trade (Hawkes, 2006; Hawkes &Murphy,
2010). Between 1990 and 2010, diets changed in contradictory ways, with some dietary patterns showing
higher consumption of healthy foods and others based on the greater intake of unhealthy foods (Imamura
et al., 2015). Middle-income countries, of which Ghana is part, embody this ambivalence by showing the
worst shifts in terms of increased consumption of unhealthy foods, but also improvements in consumption
of healthy foods (Ibid.). In Ghana, the market value of packaged food3 tripled between 2012 and 2018 and
is forecasted to grow at one of the highest rates worldwide; currently, snacksmake up one-third of the total
market value for packaged foods (Euromonitor International 2019). Per-capita expenditure on packaged
food almost doubled since 2004 (160 USD in 2018) and is now remarkably higher than in Nigeria (32
USD) (Ibid.).
In 2010, Ghana attained lower middle-income country status (Jerven, 2016) and is considered by
many as a success story given sustained GDP growth and good performance on human development
indicators (Cooke, Hague, & McKay, 2016). This has contributed to the emergence, or expansion, of
the so-called middle class (Luckham, Gyimah-Boadi, Ahadzie, & Boateng, 2005). However, as Lentz
(2015) describes, much of the Ghanaian middle class laments it cannot make ends meet. The story of
success has come under serious threat due to the collapse in export revenues for oil, cocoa and gold in
recent years, and a worsening fiscal position.4 This was accompanied by an increase in prices of fuel,
water and electricity paid by consumers, resulting in malcontent and urban protests. Economic
downturn and the rising cost of living are concerning especially for the poorer population, and for
4 S. Stevano et al.
the impact on poverty and inequality. While poverty rates have fallen, inequality in Ghana has been
on the rise since the early 1990s, as the most vulnerable social class continues to expand much more
strongly than other social groups in 2012–2017 (Agyire-Tettey et al., 2018; Annim, Mariwah, &
Sebu, 2012; Cooke et al., 2016; Euromonitor International 2019; Obeng-Odoom, 2012).
The Greater Accra Region shows the fastest-growing consumer expenditure as well as the lowest
levels of poverty and intra-regional inequality nationally, with poverty incidence at 2.6 per cent and
a GINI coefficient of 35.4 (Euromonitor International 2019; GSS, 2015). However, these statistics
hide the dynamics of inequality that are visible only at the micro-level, beyond the level of the region
or district. For instance, Obeng-Odoom (2012) suggests that inequality in Accra has been driven by
in-migration of poorer people from other urban and rural areas. Studies of the housing market in
Accra highlight precarious conditions, with rents gradually increasing and a lump-sum payment
system of one or more years of advance rent that is particularly daunting for the poor (Arku,
Luginaah, & Mkandawire, 2012). Earlier studies of livelihoods in Accra find that many households
are vulnerable due to low-paid jobs and informalised, unsecure employment (Maxwell et al., 2000).
Thus, low levels of poverty in the Greater Accra Region, relative to other regions, obscure intra-
regional and intra-urban vulnerabilities.
Against this background, we will seek to integrate the analysis of food security and consumption in
the functioning of urban food systems. Debates on urban food security in the late 1990s, as illustrated
by Maxwell (1999), were concerned with making urban food security visible as a political issue in the
context of urban poverty. The emphasis on urban poverty was renewed in the aftermath of the 2007
food crisis, when the food price shock brought to light the fragility of food security, especially for the
urban poor. While urban poverty continues to need attention, due to rapid urbanisation and scarcity of
decent employment across urban areas in the Global South (Floro & Swain, 2013), we argue that it is
now also necessary to look at the urban food question in the context of dietary change and changing
food quality. It is necessary to capture the complexity of food consumption, by analysing the micro-
level dynamics of urban food sourcing practices and food consumption. This overview provides the
broader context in which our micro-level analysis of urban food security and food consumption
among schoolchildren is to be situated.
3. Study setting and methods
This study explores vulnerability through the lens of inequality in food consumption among school-
children in Accra. The findings presented in this paper are based on triangulation of primary and
secondary data, which were collected and analysed between 2015 and 2017. Primary data collection
consisted of a student survey and qualitative focus groups with schoolchildren in five Junior High
Schools (JHS) in the Accra Metropolitan Area. In addition, throughout field research, semi-structured
interviews were conducted with representatives of key governmental and non-governmental organisa-
tions as well as the food industry. A summary of the methods for primary data collection is provided in
the Table 1, and the methodology used for interviewing schoolchildren is described in turn.
3.1. Selection of schools, student survey and focus groups
Having gained permission from the Ghana Education Service (GES) to conduct research activities in
JHS, we selected four schools in the Accra Metropolitan Area (AMA). Using key informant inter-
views and criteria such as school type (i.e. public or private), location, indicators of minimum
teaching equipment and school’s infrastructure,5 the selection aimed to choose schools attended by
children from different socio-economic backgrounds. Table 2 illustrates our selection, which repre-
sents a wide range, from a public school attended by many children in Nima, a lower-class
neighbourhood in Accra (Aguda, 2009), to a top private school charging tuition fees in USD.6
Respondents were selected randomly and the process was facilitated by teachers in each school.7
We planned to conduct at least 30 structured interviews in each school. However, we could not
The urban food question in Ghana 5
Table 1. Summary of methods for primary data collection
Phase Method Purpose
I Student survey on food consumption and nutrition knowledge To gather information on
(sample = 139 schoolchildren enrolled in JHS) children’s
● Food security
● Dietary Diversity
● Food preferences and
nutrition knowledge
● Food consumption
II Focus groups with sub-groups of survey respondents To deepen understanding of
● Intra-urban mobility and
food acquisition
● Exposure to food adver-
tising and food
preferences
● Nutrition knowledge and
food consumption
Cross-cutting Semi-structured interviews with key stakeholders (ministries, To gather information on
international organisations, food industry, school and street
food vendors) ● Children’s food
environment
● Key stakeholders’ rela-
tions to nutrition
narratives
● Food and nutrition policy
Source: Created by authors.
Table 2. Key characteristics of four selected JHS in Accra Metropolitan Area
Kanda Estate Osu Association International School
School Clustera Presbyb Jack & Jill (AIS)
Location Nima/Kanda Osu Roman Airport Residential Area
Ridge
Type Public Public Private Private
Student-teacher ratio 11.8 15.7 13.6 5.4
Access to functioning library No Yes (Girls) Yes Yes
No
(Boys)
Access to functioning No No Yes Yes
computer lab
Minimum teaching equipment Yes Yes Yes Yes
Minimum school No No Yes Yes
infrastructure
Source: Compiled by authors.
Notes: aKanda Cluster includes four schools sharing the same courtyard, these are: Ring Road East, Kanda 1,
Kanda 5 and Kanda A.M.A. Data on student–teacher ratio are given as the average across the four schools. bOsu
Presby includes Osu Presby Girls and Osu Presby Boys. Data on student–teacher ratio are given as the average
between the two schools.
complete all interviews at AIS and therefore added a fifth school, Tema International School (TIS),
which is also a top private and boarding school. The overall sample is as reported in the Table 3.
The questionnaire was composed of the following components: basic demographic information,
child’s assets, food security questions, child dietary diversity score, food knowledge and consumption
questions. We measure children’s diets through indicators of food security, dietary diversity and food
6 S. Stevano et al.
Table 3. Student survey sample
School No. of students interviewed
Osu Presby Boys and Girls 32
Kanda Estate Cluster 30
Jack & Jill 31
AIS 16
TIS 30
Total 139
Source: Compiled by authors.
consumption, and link these to food provisioning practices, which told us more about the structural
features of the urban food system.
Based on a preliminary analysis of survey data, a sub-sample of 63 schoolchildren across the four
schools in AMA were selected to take part in a series of qualitative focus groups grouped on three
themes. Focus group 1 looked at patterns of intra-urban mobility and food acquisition practices.
Focus group 2 explored the theme of children’s exposure to food advertising and food preferences.
Focus group 3 investigated children’s food and nutrition knowledge, trying to discern whether
knowledge shapes consumption.
3.2. Child asset index
The literature on child-based interviews discusses the challenges ofmeasuring child poverty as childrenmay
not have good knowledge of household income, consumption or assets (Gordon, Nandy, Pantazis,
Pemberton, & Townsend, 2003; Hannum, Liu, & Alvarado-Urbina, 2017). One route is to collect data on
children’s assets, which minimises measurement errors (Ibid.). In this approach, an indicator based on
children’s assets is a proxy of underlying household welfare (i.e. we assume that the child asset index varies
with underlying household wealth), rather than the choice of asset that directly produces welfare (c.f.
Johnston & Abreu, 2016). As such this is not a measure of child deprivation but a measure that correlates
with household wealth.We collected data on a set of assets that children have direct access to – books, tablet,
computer, dvd player, smartphone, video game console and own bedroom – and constructed a child’s asset
index using principal component analysis (PCA) (Abreu 2012; Filmer & Pritchett, 2001; Wall & Johnston,
2008). The sample was then grouped in wealth quintiles based on the asset index.
3.3. Food and diet indicators
We use four key indicators to analyse food consumption and diets: one on access to food, two on diet
quality and one on food knowledge. First, to assess food security we follow the approach used by the
WHO Global School-Based Student Health Survey (WHO, 2012), which consists of two questions:
one about going hungry and one on having breakfast, with associated frequencies. Second, as proxies
for diet quality, we use a standard dietary diversity indicator, based on 24 h recall (Arimond et al.,
2010; FAO, 2010; Ruel, 2002) and, in addition, an adapted version of the World Food Programme’s
(WFP) Food Consumption Score (FCS) (WFP, 2008), which captures the consumption frequency of
particular food items in the seven days before the interview.8 Finally, we considered food knowledge,
as food and nutrition knowledge is often a core component of nutrition interventions targeting
schoolchildren or mothers (Adelman, Gilligan, & Lehrer, 2008; Imdad, Yawar Yakoob, & Bhutta,
2011), and defined it as an awareness of key healthy foods, such as fruit and vegetables, or basic
nutrition facts, such as the importance of balanced diets and nutrient intake.
A central motivation of our analysis is to explore the association between dietary practices and
socio-economic status, as measured by the child asset index. Thus, we explore associations between
food indicators and wealth and conduct a basic regression analysis to sketch how age, gender, wealth
The urban food question in Ghana 7
and food knowledge may be associated with measures of diet quality. However, as this analysis alone
is limited, we bring together qualitative insights on the food environment to reach a fuller under-
standing of food inequalities.
4. Results
4.1. Sites and means of schoolchildren’s urban food provisioning
The school is a central and heterogeneous site for children’s school provisioning. The private schools in
our study have a canteen. In AIS, the canteen is run by a catering company and has a monthly menu with
options selected in advance by children or their families. The canteen meal includes fruit or dessert, and
vegetables as sides. Children can also buy snacks at break times from the canteen. In Jack & Jill,
however, children can buy snacks and drinks from vendors on the school’s premises, and in their canteen,
children buy food by piece, not by meal. Thus, the practice of pricing food by item, rather than by meal,
implies that children buying food from the same canteen have varying types and quantities of food.
In public schools, there are no canteens, and stationary food vendors operate within the school’s
premises and in the surrounding areas. Where options are limited at school, children prefer to go to the
street food vendors in the school’s surroundings.9 Children may also buy food before entering the school
in the morning. For example, at Osu Presby, it was common among schoolchildren to buy porridge from
the stall in front of the school in the morning. Besides permanent food stalls in the proximity of schools,
public schools tend to be a target of mobile food vendors, as we will discuss in more detail later.
Street food, defined as ‘any ready to eat food or beverage sold and sometimes prepared in outdoor
public spaces by vendors or cooks, either itinerant or stationary, either on foot or from mobile outlets’
(FAO, 2016, p. 1), is central to urban children’s food provisioning, providing much of the food consumed,
particularly in public schools. Street food employs an estimated 60,000 individuals in Accra, many of
whom are women, and its ubiquitous presence has directed attention to issues of food hygiene and safety
(Ackah et al., 2011; FAO, 2016). A high number of children in the sample, 72.6 per cent, had purchased
street food at least once in the week before the interview. Interestingly, street food is especially important
for children from poorer backgrounds, with 60 per cent and 75 per cent of children in the two poorest
quintiles, respectively, who had bought street food more than once in the previous week.
How do children purchase food? Most children in the sample, everyone except from children at
AIS and TIS, buy food for themselves with chop money. This is money handed out on a daily basis,
unless the families cannot afford to do so daily,10 and serves the specific purpose of buying food, and
in some cases for other expenses, such as transportation. The amount of chop money ranges from 1 to
2 GHS to 10 GHS per day, the average is 5 GHS/day. In contrast, children at AIS receive pocket
money, which is handed out mostly weekly and not used to buy meals. Meals are paid for by families
in advance and therefore children can use pocket money to buy some snacks. In focus groups at AIS,
children explained that they use pocket money to buy food when they go to shopping malls with
friends at weekends. This is a substantial difference for children from different socio-economic
backgrounds and one that has direct implications on the quantity and quality of food consumed.
4.2. Food security
The majority of children in the sample, 76 per cent, have not gone hungry due to lack of food in the
month before the interview. However, a remarkable 24 per cent of the sample reported that they felt
hungry once or a few times due to lack of food. In a context characterised by the lowest levels of
undernutrition in the country, this finding stands out. This result shows that the inability to access
sufficient quantities of food does occur among schoolchildren in Accra, thus suggesting that urban
food insecurity is an issue that needs attention.11
Who are the food insecure children? Looking at the association between food insecurity and
wealth, we observe that most food insecure children are in the poorer quintiles. We find a negative
8 S. Stevano et al.
and statistically significant association (p-value = 0.065) between food insecurity and wealth. The
question on breakfast reveals that the majority of children interviewed, 68 per cent, have breakfast
before going to school. However, a significant 32 per cent of children in the sample go to school
without breakfast. Many of these children buy breakfast on their way to school, from street food
vendors, or at school from the school canteen or on-site vendors. With the focus groups, we explored
some of the reasons for not having breakfast at home. Children mentioned a variety of factors that
may prima facie appear to reflect a family’s practices and habits. For example, some children said
they have to wake up very early to arrive at school on time and therefore indicated lack of time as the
reason for not having breakfast at home, others said their parents do not prepare breakfast at home
and instead give them money to buy breakfast on their way to school. However, we also find
a negative association (p-value = 0.000) between not having breakfast before going to school and
wealth, with the majority of children in this group in the two poorest quintiles.
This finding suggests that reasons mentioned for not having breakfast may conceal an underlying
link with socio-economic status. Having to wake up very early in the morning may reflect the
location and housing conditions of children’s homes, how they go to school, their parents’ occupa-
tions, and their roles within their households. For example, we find that 93.5 per cent of children in
public schools walk to school while this percentage drops to only 10.4 per cent of children in private
schools, as more than half of them are driven to schools. These patterns of intra-urban mobility shape
where children acquire food, with food vendors concentrating in the proximity of public schools.
With regard to food security, a key concern is the timing of the first meal. On the day before the
interview 53.4 per cent of the sample had breakfast by 7:30 while the rest of the sample had their first
meal between 8:00 and 13:00. There was a group of children, 15 per cent, who had their first meal at
the first school break, around 10:00 or later. Most of these children are in the two poorest wealth
quintiles. This may reflect the lack of resources to have two meals in the morning, resulting in the
practice of postponing breakfast.
4.3. Diet quality
The dietary diversity indicator captures high consumption for cereals (99.3%), oils and fats (65.5%),
sweets (50.4%), meat (48.2%), vegetables (48.2%) and dairy products (43.9%). Instead, very low
levels of consumption were recorded for vitamin-A-rich fruit and vegetables (3.6% and 9.4%,
respectively) and nuts and seeds (9.4%). We find a positive and statistically significant association
between dietary diversity and wealth (p-value = 0.007).
The FCS is associated positively with wealth, with a high level of statistical significance (p-value =
0.000). This suggests that children in wealthier quintiles consume healthier foods. In particular, we
see that dairy and vegetables consumption increases in frequency with wealth. However, for other
food items, the association with wealth is not clear-cut. In particular, this appears to be the case for
foods and drinks that are consumed as snacks. Consumption of packaged snacks and carbonated
drinks appears to cut across wealth quintiles, with poorer and wealthier children reporting similar
levels of consumption. We argue that these forms of food consumption reflect the widespread
availability and relative affordability of these foods. Packaged snacks and soft drinks are sold by
food vendors on the schools’ premises or in surrounding areas. Prices are such that most children can
afford to buy these foods, at least occasionally. Indeed, packaged snacks and soft drinks are among
the cheapest foods at school or on the road.
To provide an example, we briefly discuss FanMilk snacks. FanMilk is a multinational that has
been present in Ghana since the 1960s. FanMilk’s distribution structure relies on thousands of mobile
street vendors operating particularly in urban areas.12 Its main products are a range of packaged
frozen snacks, which may be yoghurt-based, chocolate-based or have a fruit flavour. In interviews
conducted with FanMilk vendors, it emerged that many vendors go to public schools at break times
and at the end of the schoolday. According to the vendors, FanMilk snacks are popular among
children, which is confirmed by 72.7 per cent of children in our sample reporting having eaten
The urban food question in Ghana 9
FanMilk snacks in the previous week, with the highest consumption frequency found in the middle
wealth quintiles. Children use their chop or pocket money to buy these snacks that are widely
available and relatively affordable. These snacks are less available to children in top private schools
because food vendors are not allowed to operate on the school’s premises and proximity. Conversely,
children from the poorest quintile can easily find these snacks at school but the lower frequency
reflects smaller amounts of chop money. At times, vendors cut the snacks in half so that two children
can share the cost. This example provides a glimpse on the functioning of the urban food system
where food companies make use of informal channels to boost sales of affordable and widely
available packaged foods.
4.4. Food knowledge
The majority of children in the sample (67.7%) have basic food knowledge. The association between
food knowledge and wealth is positive but statistically insignificant (p-value = 0.107). It is however
unlikely that the association is driven by the quality of nutrition education at school, as children in
AIS and TIS, over-represented in the wealthiest quintile, do not have home science classes while all
other children in the sample do. For example, in Osu Presby Girls, children were well aware of the
importance of checking expiry dates and labels of purchased food as they were taught so by their
home science teacher. In contrast, with the absence of home science class, it is more likely that
children in the wealthier quintiles are more exposed to food and nutrition messages within their
families. In focus groups, it emerged that children learn about nutrition in a range of ways: at school,
at home, through their families, or by watching television.
In focus groups, children often mentioned the importance of hygiene. Food safety is a priority in
the national nutrition policy 2013–2017 (Ministry of Food and Agriculture [MOH], 2013), and has
been incorporated into the teaching curriculum. The Ghanaian government, has paid attention to this
issue, especially in relation to street food (FAO, 2016; MOH, 2013). In some interviews, children said
that their mothers did not allow them to buy street food due to food safety concerns. Importantly, the
focus groups participants in public schools believed that the food available to them during school
time was not hygienic but the lack of alternative options meant that they nonetheless buy it.
In the survey and in the focus groups, children uniformly reported that they have healthy foods at home
and unhealthy foods at school or in the street, reflecting views on food quality. Nevertheless, there are
often gaps between taste and food knowledge, with many children mentioning fast foods as their favourite
ones. Does nutrition knowledge inform food consumption then? Linear regression analysis suggests that
food knowledge is positively associated with dietary diversity but the relationship with food consumption
score is insignificant. The results reported in the table below suggest that wealth has a stronger explana-
tory power for both dietary diversity and food consumption. Food consumption appears to be negatively
associated with the child’s age, which is a result that is difficult to interpret (Table 4).
Linear regression analysis does not illuminate important aspects of the relationship. In quali-
tative interviews, children said that they select foods based on taste and desirability, even if they
are unhealthy. Therefore, accessibility, affordability and desirability of food are critical interacting
factors. On visits to public schools, we occasionally found that all interviewed children had the
same food for lunch because they all bought it from a convenient food stall. For children who buy
food with chop money, the food needs to be affordable and desirable. We explored children’s
exposure to food advertisement and found that most children could easily recall food adverts seen
on television and on billboards. Many of them were also aware that some adverts target children
specifically, using messages that appeal to children, such as the promise of increased energy for
sport, catchy colours and memorable jingles. Interestingly, many children doubted of the relia-
bility of advertisements and were particularly sceptical when foods are publicised as healthy and
nutritious.
10 S. Stevano et al.
Table 4. Linear regression analysis
Child dietary diversity Child food consumption score
b SE B SE
Age −0.118 (0.15) −3.247*** (0.94)
Gender 0.615 (0.39) 0.404 (2.43)
Wealth 1.621*** (0.51) 12.267*** (3.14)
Food knowledge 0.948* (0.43) 2.35 (2.66)
Constant 1.331 (1.83) 34.083*** (11.35)
R2 0.120 0.256
Observations 139 139
Source: Own analysis from survey data.
Notes: *p < 0.10; **p < 0.05; ***p < 0.01.
5. Discussion and conclusions: the urban food security question today
The results extend several debates on urban food consumption. First, the evidence shows inequality in
food quality and food consumption partly driven by socio-economic status. Children from poorer
backgrounds are more vulnerable to food insecurity, lower dietary diversity, and poor food consump-
tion. Thus, food insecurity, far from being only a rural problem, concerns the urban poor too, and it
does so even in the region with the lowest levels of undernutrition nationwide. Low or irregular chop
money, which is most likely linked to low and fluctuating household incomes, poses a major obstacle
to the ability of children to meet their food needs during the school day, and a group of children in the
sample have their first meal at the first or second break during a school day.
Although it is clear that urban food consumption inequality is driven by wealth, the dynamics are
complex. The consumption of packaged and processed foods (among the culprits of rising obesity
and associated health problems, see Monteiro et al., 2013) is not only accessible to the wealthier
groups of the population, but cuts across wealth groups. These foods tend to be desirable, accessible
and relatively affordable. The use of informal distribution channels for packaged foods means they
can be bought from street vendors at school and on the roadside, and not only at supermarkets,
ensuring their accessibility. With regard to affordability, one comprehensive study on food prices in
middle-income countries found that, with the food industry being able to contain costs by using cheap
ingredients, the cost of processed foods has either stagnated or fallen while the price of fruit and
vegetables has increased over time (Wiggins & Keats, 2015).
In the 1990s, the question of urban food security was seen as embedded in that of urban poverty
(Maxwell, 1999). We argue that the urban food question today is defined by two intersecting
phenomena: urban inequality and global dietary change. Thus, while the urban poor continue to
face the fundamental challenge of adequate food access, a broader definition of urban food security
recognises the political economy of food quality and the food industry’s role in creating a food
environment that provides consumers with unhealthy food options that are widely available, cheap
and enticing. Therefore, agricultural and food policies that seek to ensure the availability of afford-
able staples for growing urban population through are no longer sufficient. However, Masters et al.
(2018) show that food policy across African countries continues to pursue lower food prices. The
urban food security question today can be addressed through agricultural, industrial and trade policies
that regulate the production of cheap, processed, unhealthy foods.
We recognise that our findings are limited by a narrow focus on schoolchildren and more research
is needed to grasp the complexity of urban food systems. We also hope that our paper stimulates
a debate on the methodological advancements needed to connect micro-level food consumption and
macro-level analysis of food systems. We conclude by highlighting two lacunae where future
research and policy should expand into to enhance our ability to respond to the question of urban
The urban food question in Ghana 11
food consumption amidst growing inequality and potentially harmful dietary change. First, scholars
and policymakers are often guilty of overlooking food processing (Scrinis, 2013). This is important as
the urban food question and urban poverty are not restricted to the issue of food access but also
encompass issues of diet quality. Second, while policies that can radically improve food environ-
ments are both necessary but complex, enhancing children’s food environment at school may be more
straightforward. In Ghana, there is currently no regulation on the types of food available to children at
school. Here the debates on food desert in the North can bring useful insights to policymakers hoping
to improve children’s food consumption, with potential long-lasting benefits for them as future adults.
Acknowledgements
This research has been made possible by the people who took part in the research activities in Ghana,
in particular, the students of the schools in Accra and Tema. We acknowledge the work of Millicent
Awuku, who worked with us as a research assistant while doing her PhD in Economics at the
University of Ghana, Legon, and are grateful to the comments received by Dr Fiorella Picchioni, the
anonymous referee and the editor. This research was funded by the Danish Strategic Research
Council. The data are available from the authors upon request.
Funding
This work was supported by the Danish Strategic Research Council [HAPFAM project].
Notes
1. The GHI is compiled by the International Food Policy Research Institute (IFPRI) combining four indicators: FAO
prevalence of undernourishment, child stunting, child wasting and child mortality. The GHI is considered to be a good
proxy of hidden hunger, linked to micronutrient imbalances.
2. It is recognised that the national statistics on food production and trade compiled by the Food and Agriculture
Organisation (FAO) are subject to errors. The quality of this data is especially low for poorer countries, where data
gaps are filled with model-based estimations, which often differ substantially from actual production and trade volumes
(Hawkesworth et al., 2010). These statistics may be useful to describe some broad patterns of food production and trade,
but given their limitations, they should only be used in combination with other evidence.
3. Packaged food includes cooking ingredients, dairy products and baby food, snacks and staple foods.
4. On 19 January 2016, Ghana Web reports that the total revenue for the three most important export commodities – oil, gold
and cocoa – has declined by 2.4 billion USD between 2014 and 2015.
5. Minimum teaching equipment and minimum school infrastructure are two World Bank’s Service Delivery Indicators
(Molina & Martin, 2015) used to assess the delivery of education and health. Minimum teaching resources captures the
availability of (i) whether a random classroom has a functioning blackboard and chalk, (ii) the share of students with pens,
and (iii) the share of students with notebooks. Minimum infrastructure resources reflects the availability of (i) functioning
toilets operationalised as being clean, private, and accessible (ii) sufficient light to read the blackboard from the back of
the classroom. The indicators, operationalised as binary, are based on researchers’ observation of selected schools.
6. Classifying schools was preferable to classifying neighbourhoods due to lack of locality data and intra-locality hetero-
geneity (Maxwell et al., 2000), as well as the fact that children going to private schools tend not to live in the school area.
7. We considered the possibility that the teacher’s participation could introduce a bias towards higher-achieving pupils.
However, respondents obtained a thank-you gift of pens and pencils, it was also possible teachers would prioritise children
from more vulnerable backgrounds. Thus, overall it is hard to identify an overall bias.
8. It is an adapted FCS because we did not collect all data to calculate the full FCS. We have information on a smaller
number of food items grouped in five food groups – fruit, vegetables, dairy, meat/fish, and sugar. Therefore, we did not
calculate a score but instead used the continuum variable based on the weightings provided by WFP (2008).
9. Most children interviewed at Kanda Estate reported that they buy food from the food vendors across the road from the
school, where there is also a minibus (trotro) station and there are more options.
10. Five children said that they do not receive chop money on a daily basis, and therefore they need to spread the available
cash over two or three days. At Kanda Estate, the teachers told us that, if there is any left-over food from the School
Feeding Programme in the primary schools (which are in the same buildings), they give it to children in JHS who struggle
to buy their own food.
12 S. Stevano et al.
11. We note that the question has a subjective character and may have captured, in some cases, feeling hungry due to
circumstantial lack of food rather than real lack of resources to acquire food.
12. The exact number of FanMilk street vendors is unknown because they are not employed by FanMilk, but considered to be
independent vendors, fully operating in the informal economy. The structure of employment is similar to that becoming
prominent in the gig economy: vendors are given equipment (so they are recognisable on the streets) and training by
FanMilk but they are nonetheless treated as self-employed.
ORCID
Deborah Johnston http://orcid.org/0000-0003-0065-1105
References
Abreu, A. J. G. (2012) Migration and development in contemporary Guinea-Buissau: A political economy approach (PhD
thesis in Economics). SOAS University of London.
Ackah, M., Gyamfi, E. T., Anim, A. K., Osei, J., Hansen, J. K., & Agyemang, O. (2011). Socio-Economic profile, knowledge
of hygiene and food safety practices among street-food vendors in some parts of Accra-Ghana. Internet Journal of Food
Safety, 13, 191–197.
Adelman, S., Gilligan, D., & Lehrer, K. (2008). How effective are food for education programs? A critical assessment of the
evidence from developing countries (Vol. 9). Washington D.C.: International Food Policy Research Institute.
Aguda, N. D. (2009). Bringing food home: A study on the changing nature of household interaction with urban food markets in
Accra, Ghana, Ontario (PhD thesis). Department of Geography, Queen’s University.
Agyire-Tettey, F., Ackah, C. G., & Asuman, D. (2018). An unconditional quantile regression based decomposition of spatial
welfare inequalities in Ghana. The Journal of Development Studies, 54(3), 537–556.
Amanor, K. S. (2015). Rising powers and rice in Ghana: China, Brazil and African agricultural development. Working Paper,
Future Agricultures, (123).
Amanor, K. S., & Chichava, S. (2016). South–South cooperation, agribusiness, and African agricultural development: Brazil
and China in Ghana and Mozambique. World Development, 81, 13–23.
Annim, S. K., Mariwah, S., & Sebu, J. (2012). Spatial inequality and household poverty in Ghana. Economic Systems, 36(4),
487–505.
Arimond, M., Wiesmann, D., Becquey, E., Carriquiry, A., Daniels, M. C., Deitchler, M., … Torheim, L. E. (2010). Simple food
group diversity indicators predict micronutrient adequacy of women’s diets in 5 diverse, resource-poor settings. The
Journal of Nutrition, 140(11), 2059S–2069S.
Arku, G., Luginaah, I., & Mkandawire, P. (2012). “You either pay more advance rent or you move out”: Landlords/Ladies’ and
Tenants’ Dilemmas in the low-income housing market in Accra, Ghana. Urban Studies, 49(14), 3177–3193.
Cooke, E., Hague, S., & McKay, A. (2016). The Ghana poverty and inequality report: Using the 6th Ghana Living Standards
Survey. March 2016.
Cudjoe, G., Breisinger, C., & Diao, X. (2010). Local impacts of a global crisis: Food price transmission, consumer welfare and
poverty in Ghana. Food Policy, 35(4), 294–302.
Demographic and Health Survey (DHS). (2014). Ghana, Ghana Statistical Service (GSS), Ghana Health Service (GHS), and
ICF International. 2015. Rockville, MD: GSS, GHS, and ICF International.
Euromonitor International (2019). Passport database, accessed in June 2019.
FAOSTAT http://www.fao.org/faostat/en/#data accessed in April 2017
Filmer, D., & Pritchett, L. (2001). Estimating wealth effects without expenditure data – Or tears: An application to educational
enrolments in states of India. Demography, 38(1), 115–132.
Floro, M. S., & Swain, R. B. (2013). Food security, gender, and occupational choice among urban low-income households.
World Development, 42, 89–99.
Food and Agriculture Organisation (FAO). (2010). Guidelines for measuring individual and household dietary diversity.
Rome: Food and Agriculture Organisation.
Food and Agriculture Organisation (FAO). (2015). The state of agricultural commodity markets. Trade and food security:
Achieving a better balance between national priorities and the collective good. Rome: Food and Agriculture Organisation.
Food and Agriculture Organisation (FAO). (2016). Street food in Urban Ghana. A desktop review and analysis of findings and
recommendations from existing literature. Accra: Food and Agriculture Organisation.
GHI (2016). Global Hunger Index 2016. Getting to zero hunger, International food Policy Research Institute (IFPRI) and
Concern Worldwide.
Girdner, J., Olorunsola, V., Froning, M., & Hansen, E. (1980). Ghana’s agricultural food policy: Operation feed yourself. Food
Policy, 5(1), 14–25.
Gordon, D., Nandy, S., Pantazis, C., Pemberton, S., & Townsend, P. (2003). Child poverty in the developing world. Bristol:
Policy Press.
The urban food question in Ghana 13
GSS (2015). Ghana poverty mapping report. Ghana Statistical Service, 2015.
Hannum, E., Liu, R., & Alvarado-Urbina, A. (2017). Evolving approaches to the study of childhood poverty and education.
Comparative Education, 53(1), 81–114.
Hawkes, C. (2006). Uneven dietary development: Linking the policies and processes of globalization with the nutrition
transition, obesity and diet-related chronic diseases. Globalization and Health, 2(1), 4.
Hawkes, C., & Murphy, S. (2010). An overview of global food trade. In C. Hawkes, C. Blouin, S. Henson, N. Drager, &
L. Dubé (Eds.), Trade, food, diet and health: Perspectives and policy options (pp. 16–32). Oxford: John Wiley & Sons.
Hawkes, C., Blouin, C., Henson, S., Drager, N., & Dubé, L. (eds). (2010). Trade, food, diet and health: Perspectives and policy
options. Oxford: John Wiley & Sons.
Hawkesworth, S., Dangour, A. D., Johnston, D., Lock, K., Poole, N., Rushton, J., … Waage, J. (2010). Feeding the world
healthily: The challenge of measuring the effects of agriculture on health. Philosophical Transactions of the Royal Society
of London B: Biological Sciences, 365(1554), 3083–3097.
Imamura, F., Micha, R., Khatibzadeh, S., Fahimi, S., Shi, P., Powles, J., & Mozaffarian, D. (2015). Dietary quality among men
and women in 187 countries in 1990 and 2010: A systematic assessment. The Lancet Global Health, 3(3), e132–e142.
Imdad, A., Yawar Yakoob, M., & Bhutta, Z. A. (2011). Impact of maternal education about complementary feeding and
provision of complementary foods on child growth in developing countries. BMC Public Health, 11(Suppl. 3), S25.
Jerven, M. (2016). Research note: Africa by numbers: Reviewing the database approach to studying African economies.
African Affairs, 115(459), 342–358.
Johnston, D., & Abreu, A. (2016). The asset debates: How (not) to use asset indices to measure well-being and the middle class
in Africa. African Affairs, adw019.
Kollamparambil, U. (2017). Impact of internal in-migration on income inequality in receiving areas: A district level study of
South Africa. The Journal of Development Studies, 53(12), 2145–2163.
Lagerkvist, C. J., Hess, S., Okello, J., & Karanja, N. (2013). Consumer willingness to pay for safer vegetables in urban markets
of a developing country: The case of Kale in Nairobi, Kenya. The Journal of Development Studies, 49(3), 365–382.
Lentz, C. (2015). Elites or middle classes?: Lessons from transnational research for the study of social stratification in Africa.
Working Papers of the Department of Anthropology and African Studies of the Johannes Gutenberg University Mainz 161.
Lerner, A., Sweeney, S., & Eakin, H. (2014). Growing buildings in corn fields: Urban expansion and the persistence of maize
in the Toluca Metropolitan Area, Mexico. Urban Studies, 51(10), 2185–2201.
Levin, C. E., Ruel, M. T., Morris, S. S., Maxwell, D. G., Armar-Klemesu, M., & Ahiadeke, C. (1999). Working women in an
urban setting: Traders, vendors and food security in Accra. World Development, 27(11), 1977–1991.
Luckham, R., Gyimah-Boadi, E., Ahadzie, W., & Boateng, N. (2005). The middle classes and their role in national
development. CDD/ODI Policy Brief No. 3, November 2005.
Masters, W. A., Rosenblum, N. Z., & Alemu, R. G. (2018). Agricultural transformation, nutrition transition and food policy in
Africa: Preston curves reveal new stylised facts. The Journal of Development Studies, 54(5), 788–802.
Maxwell, D. (1999). The political economy of urban food security in Sub-Saharan Africa. World Development, 27(11),
1939–1953.
Maxwell, D., Levin, C., Armar-Klemesu, M., Ruel, M., Morris, S., & Ahiadeke, C. (2000). Urban livelihoods and food and
nutrition security in Greater Accra, Ghana. Washington, DC: International Food Policy Research Institute.
Ministry of Food and Agriculture (MOH). (2013). National Nutrition Policy 2013-2017. Accra: Ministry of Health.
Molina, E., & Martin, G. (2015). Education service delivery in Mozambique. World Bank, October 2015.
Monteiro, C. A., Moubarac, J. C., Cannon, G., Ng, S. W., & Popkin, B. (2013). Ultra-processed products are becoming
dominant in the global food system. Obesity Reviews, 14(S2), 21–28.
Nyanteng, V. K., & Wayo Seini, A. (2000). Agricultural policy and the impact on growth and productivity in 1970-1995. In
E. Aryeetey, J. Harrigan, & M. Nissanke (Eds.), Economic reforms in Ghana: The miracle and the mirage (pp. 267–283).
Africa World Press.
Obeng-Odoom, F. (2012). Neoliberalism and the urban economy in Ghana: Urban employment, inequality, and poverty.
Growth and Change, 43(1), 85–109.
Ofori-Asenso, R., Agyeman, A. A., Laar, A., & Boateng, D. (2016). Overweight and obesity epidemic in Ghana—A systematic
review and meta-analysis. BMC Public Health, 16(1), 1239.
Ouma, S. (2015). Export markets: The making and unmaking of global food connection in West Africa. Chichester, UK: John Wiley.
Popkin, B. M., Adair, L. S., & Ng, S. W. (2012). Global nutrition transition and the pandemic of obesity in developing
countries. Nutrition Reviews, 70(1), 3–21.
Ruel, M. T. (2002) Is dietary diversity an indicator of food security or dietary quality? A review of measurement issues and
research needs, FCND Discussion 319.
Scrinis, G. (2013). Nutritionism: The science and politics of dietary advice. New York: Columbia University Press.
Wall, M., & Johnston, D. (2008). Counting heads or counting televisions: Can asset-based measures of welfare assist policy-
makers in Russia? Journal of Human Development, 9(1), 131–147.
WFP. (2008). Food consumption analysis calculation and use of the food consumption score in food security analysis. Rome:
World Food Programme.
WHO (2012). Global school-based student health survey. Junior High Schools Report, World Health Organisation.
Wiggins, S., & Keats, S. (2015). The rising cost of a healthy diet changing relative prices of foods in high-income and
emerging economies. London: Overseas Development Institute (ODI).