Acknowledgements We are grateful to the Central Statistical Agency for providing access to datasets used in the analysis. We thank Kebede Tafesse, Victor Pinga, Dr. Girma Abebe and Regional NSA Advisors, namely Dires Beza, Haimanot Woldemariam, Habtu Assefa, Muluken Tamene, Alemayehu Lebeta, and Muhyadin Mohammed as well as the colleagues at the Food and Nutrition Coordination Office at the Ministry of Agriculture for useful comments on the earlier versions of this document. Alive & Thrive is an initiative to save lives, prevent illness, and ensure healthy growth and development through improved breastfeeding and complementary feeding practices. Good nutrition in the first 1,000 days, from conception to 2 years of age, is critical to enable all children to lead healthier and more productive lives. Alive & Thrive is scaling up nutrition through large-scale programs in several countries in Asia and Africa and through strategic technical support and the dissemination of innovations, tools, and lessons worldwide. Alive & Thrive is funded by the Bill & Melinda Gates Foundation and the governments of Canada and Ireland. The initiative is managed by FHI 360. The Alive & Thrive initiative, managed by FHI 360, is currently funded by the Bill & Melinda Gates Foundation, Irish Aid, the Tanoto Foundation, and UNICEF. Disclaimer This publication has not been peer reviewed. Any opinions stated in this publication are those of the author(s) and are not necessarily representative of or endorsed by IFPRI. Suggested Citation Hirvonen K, and Wolle A. Consumption, production, market access and affordability of nutritious foods in the Afar Region of Ethiopia. Addis Ababa, Ethiopia: Alive & Thrive and International Food Policy Research Institute, 2019. Alive & Thrive FHI 360 1825 Connecticut Avenue NW Washington, DC 20009 aliveandthrive@fhi360.org www.aliveandthrive.org mailto:aliveandthrive@fhi360.org http://www.aliveandthrive.org/ http://www.aliveandthrive.org/ 2 Table of contents 1. Introduction .......................................................................................................................... 3 2. Description of the data sources used in this report .............................................................. 5 2.1. Demographic and Health Survey (DHS) data .............................................................. 5 2.2. PSNP evaluation survey ............................................................................................... 5 2.3. CSA agricultural sample survey data ........................................................................... 6 2.4. CSA retail price survey data ......................................................................................... 7 3. Consumption of nutritious foods ......................................................................................... 7 3.1. Child diets ..................................................................................................................... 7 3.2. Household diets .......................................................................................................... 12 4. Food production from a nutrition perspective ................................................................... 14 5. Market availability of nutritious foods .............................................................................. 16 6. Affordability of nutritious foods ........................................................................................ 20 7. Conclusions ........................................................................................................................ 25 7.1. Legumes and nuts ....................................................................................................... 25 7.2. Dairy products ............................................................................................................ 25 7.3. Flesh foods .................................................................................................................. 26 7.4. Eggs ............................................................................................................................ 26 7.5. Vitamin A-rich fruits and vegetables .......................................................................... 26 7.6. Other fruits and vegetables ......................................................................................... 26 7.7. Limitations .................................................................................................................. 27 References ................................................................................................................................. 28 Appendix A: Recalculating affordability considering the second cheapest food items in fruit and vegetable categories ........................................................................................................... 32 3 1. Introduction Poor dietary quality is a significant risk factor for stunting and micronutrient deficiencies among young children and globally one of the leading causes of premature death and disease (Arimond & Ruel, 2004; Forouzanfar et al., 2015). Dietary quality is typically proxied by diversity of the consumed diet. Foods with similar nutritional qualities are first grouped together and dietary diversity is measured by the number of different food groups consumed in a certain time interval. For example, the World Health Organization recommends that children 6-23 months consume at least four food groups (out of seven) every day. Based on this metric, Ethiopian children in this age range consume one of the least diversified diets in sub-Saharan Africa (Hirvonen, 2016) with only 14 percent meeting the WHO recommendation (CSA & ICF, 2016). Recent analysis of the timing of growth faltering of young children suggests that poor complementary feeding practices, including poor dietary quality, is an important risk factor for stunting in Ethiopia (Hirvonen, Headey, Golan, & Hoddinott, 2019). The available evidence suggests that diets are monotonous also at the household level. For example, in 2011, the average Ethiopian household consumed only 42 kg of fruits and vegetables in a year per adult equivalent (Hassen Worku, Dereje, Minten, & Hirvonen, 2017) – far below the World Health Organization’s recommendation of 146 kg per year (Hall, Moore, Harper, & Lynch, 2009). Such monotonous diets are likely to increase the risk of various chronic diseases. For example, a recent global study indicated elevated mortality as well as major cardiovascular disease risk when energy intake from carbohydrates exceed 60 percent (Dehghan et al., 2017). This is particularly worrying for Ethiopia as recent estimates by the Ethiopian Public Health Institute suggest that 60-80 percent of the energy intake of children and adults comes from carbohydrates (Misganaw et al., 2017). Meanwhile, cardiovascular diseases are among the top causes of premature mortality in the country (Misganaw et al., 2017). Another indication of limited dietary quality is that micronutrient deficiencies in Ethiopia are widespread. Nearly 60 percent of young children are anemic (CSA & ICF, 2016) and more than one-third suffer from Vitamin A deficiency (Demissie, Ali, Mekonen, Haider, & Umeta, 2010). Recent research from Ethiopia further highlights the importance of fruit and vegetable consumption among pregnant and lactating women. Data from health centers in rural Oromia, Zerfu, Pinto, and Baye (2018) find 4 that pregnant women who consumed fruits and dark green leafy vegetables more frequently were less likely to experience adverse pregnancy outcomes. Moreover, analyzing samples of breastmilk from mothers in rural Amhara, Abebe, Haki, Schweigert, Henkel, and Baye (2018) find very low concentrations of Vitamin A in milk. These issues are well-acknowledged by the government of Ethiopia. The national nutrition programme sets out ambitious plans to increase the year-around availability, access, and consumption of nutritious foods (GFDRE, 2016a). Through its 'Seqota Declaration of 2015,' Ethiopia further announced its ambitious goal of ending child malnutrition by 2030 (GFDRE, 2016b). These nutrition strategies are endorsed by several ministries highlighting that the emerging consensus that combating under-nutrition and poor diets requires multi-sectoral efforts (Menon & Frongillo, 2018). At the core of these efforts is the urgent need to transform food systems in Ethiopia to support healthier diets (Gebru et al., 2018). The concept of food system captures the food environment (affordability and accessibility), consumer preferences, and the food supply system that is formed of production, storage, transportation, processing, and marketing (Scott, 2017). It is now well acknowledged that the dietary choices of individuals are determined by the surrounding food system. Against this backdrop, in this report, we aim to gain a better understanding of the food sub system1 in the Afar region of Ethiopia by analyzing affordability, accessibility, and consumption of nutritious foods. 2 The region hosts more than 1.6 million people out of which most reside in rural areas and rely on agricultural production as their main livelihood (CSA, 2018b). About 41 percent of the children under 5 in Afar are chronically under-nourished (or stunted) (Hirvonen et al., 2019) while 39 percent of the women 15-49 years of age are underweight (CSA & ICF, 2016).3 Moreover, according to the Ethiopian Public Health Institute (EPHI), 17 percent of children 6-59 months are Vitamin A deficient (EPHI, 2016). 1 We call this a sub system analysis because due to data limitations, we do not assess consumer preferences and we are also forced to overlook (due to data limitations) important parts of the food supply system: storage, transportation, processing, marketing as well as inter-regional and international trade. 2 Similar analysis was carried out for six regions in total: Afar, Amhara, Oromia, Somali, SNNP, and Tigray. 3 About 8 percent of the women in this age group in Afar are over-weight or obese. 5 This report is structured as follows. In the subsequent section, we describe the data used in this report. In section 3, we assess the consumption of nutritious foods among young children and also carry out a similar assessment at the household level. In section 4, we study the production of nutritious foods in the region. In section 5, we examine the availability of nutritious foods in rural markets. In section 6, we assess the affordability of nutritious foods in the region. Section 7 concludes and summarizes the findings. 2. Description of the data sources used in this report All the analyses in this report are based on secondary data collected by the Central Statistical Agency (CSA) of Ethiopia or the International Food Policy Research Institute (IFPRI). This section describes these data sources. 2.1. Demographic and Health Survey (DHS) data We use the 2016 Demographic and Health Survey (DHS) data for Ethiopia to analyze child diets. This survey is nationally as well as regionally representative and was implemented by the CSA with technical assistance from the ICF. The interviews took place between January 18, 2016, and June 27, 2016. A total of 16,650 households were interviewed in all regions of Ethiopia, and 1,220 households in Afar. Apart from rich information on various health outcomes and socio-economic characteristics, the DHS collects information about complementary feeding practices of young children. 2.2. PSNP evaluation survey We use the PSNP-4 evaluation data to assess household dietary diversity as well as the market availability of nutritious foods. The latest (midline) round was administered in June/July 2018 in Afar, Amhara, Oromia, Somali, SNNP, and Tigray regions in localities that benefit from the PSNP program. In Afar region, 1,079 PSNP households and poor non-PSNP households with roughly 50-50 split were interviewed. These households group into 36 communities and 12 woredas4. The household survey instrument asked yes/no questions about household's food consumption in the past seven days. These data sets are used to assess household dietary 4 These woredas are Afambo, Awash Fentale, Berhahile, Bure Mudaytyu, Chefira, Dalole, Dawe, Elidar, Golina, Hadeleala, Konaba and Teru. 6 diversity in the region. The evaluation also included a market survey; in each community, the team visited the local food market to assess food availability and prices. We use these data sets to assess the availability of nutritious foods in the region. 2.3. CSA agricultural sample survey data Building on Baye, Hirvonen, Dereje, and Remans (2019), we use the CSA's Agricultural Sample Survey reports to estimate the annual agricultural production in the country. To do so, we rely on CSA's annual flagship reports: 1) Report on Area and Production of Major Crops, private peasant holdings This report provides the total annual crop output in 10 out of the 11 administrative regions of the country. Only the production in the capital, Addis Ababa, is not reported. The crop output for meher (long rainy season and the main cropping season for most part of the country) and belg (short rainy season) are reported separately. 2) Report on Livestock and Livestock Characteristics, private peasant holdings This report provides the annual estimates of the livestock population and livestock production. The sample sizes are large – typically containing more than 40,000 rural households. These reports are based on nationally and regionally representative data collected by the CSA each year. We digitalized these reports. For crop output, we aggregated the total annual production for each crop in each region during meher and belg seasons as well as the total crop output produced by the commercial sub-sector. For livestock products, we took the total annual milk (cow and camel), eggs, and honey produced. The CSA reports give the number of different livestock types slaughtered. These numbers were converted to kilograms of beef, sheep, goat, camel, and poultry meat using conversion factors to account for edible portions only (FAO, 1972). The annual crop and livestock production was then converted into energy (kilocalories) using the Ethiopian food composition tables (EPHI, undated). We also used FAO reject estimates which correspond to weight estimations of edible portions of foods produced to calculate energy and nutrients supplied. The total energy was then divided by the total population for that specific year based on the estimates by the CSA (2013). The energy and nutrients produced were expressed on a per day basis by dividing by 365 to enable comparison with daily requirements. 7 2.4. CSA retail price survey data For the affordability analysis, we use monthly retail price data collected by the CSA. The original purpose of these data sets is to calculate the official consumer price index (CPI) in the country. These price data are collected from about 110 markets in all regions of Ethiopia with the number of markets approximately proportional to the region's size in terms of population. The CSA enumerators visit these markets every month and collect price data for more than 400 food and non-food items. For each item, the enumerators target three price quotations from different traders. For more information about this survey, see Headey, Nisrane, Worku, Dereje, and Taffesse (2012) and Bachewe, Hirvonen, Minten, and Yimer (2017). 3. Consumption of nutritious foods 3.1. Child diets Child diets are described using the World Health Organization recommended infant and young child feeding (IYCF) indicators, which are widely accepted and used to capture optimal feeding practices in populations (WHO, 2010). Here, we focus on indicators related to adequate consumption of high-quality complementary foods that are dense in key micro and macronutrients. We assess this using the protocols recommended by the World Health Organization (WHO, 2008). The DHS-2016 survey asked mothers a series of yes/no questions about foods consumed in the past 24 hours by children 6-23 months. These responses were grouped into the following categories: 1. Grains, roots, and tubers (e.g., barley, enset, maize, teff, and wheat); 2. Legumes and nuts (e.g. chickpea, beans, groundnut); 3. Dairy products (milk, yogurt, cheese); 4. Flesh foods (meat, poultry, and fish products); 5. Eggs; 6. Vitamin A-rich fruits and vegetables (e.g. mango, pumpkin, carrots); 7. Other fruits and vegetables (e.g. onions, tomatoes, bananas). This yields a count ranging in value from zero to seven. WHO (2008) recommends that children in this age range consume daily at least from four food groups out of the above seven food groups. This number was selected to maximize the likelihood that in addition to the staple food (grain, 8 root, tuber), the child consumed animal-source foods as well as fruits and vegetables in the previous day (WHO, 2008). This relatively simple indicator is highly correlated with more detailed measures of food intake (Ruel, 2003) as well as with children's micronutrient intakes5. Moreover, several studies show strong correlations with this dietary diversity score and longer term measures of children’s nutritional status (e.g. child stunting prevalence) in a number of countries including Bangladesh, Ethiopia, India, and Zambia (Jones et al., 2014). In Afar, the average 6-23-month-old child consumes from 1.3 food groups and only 2 percent of the children meet the criteria for minimum dietary diversity (at least four food groups in the last 24 hours). Children's diets are somewhat more diverse in urban areas; none (0 percent) of the rural children and 10.1 percent of the urban children in the Afar region met the minimum dietary diversity. Figure 3.1 shows the full distribution of children's dietary diversity in the region. We can use this graph to understand how far the children are from meeting the recommended dietary diversity. The bars falling short from the recommended four food groups are marked with gray color. Starting from the bottom part of the distribution, we see that a large fraction of children (22 percent) did not consume from any food groups in the previous 24 hours. These are mostly children that were still exclusively or intensively breastfed,6 indicating that the initiation of complementary feeding is often delayed in the region. Meanwhile, 34 percent of the children consumed only from one food group. For these children, the dietary diversity would have to increase by three food groups in order to meet the minimum recommended dietary diversity. About 35 percent were two food groups shy from meeting the recommendation while 7 percent of the children were missing just one food group. The remaining 2 percent consumed from four or more food groups. 5 See Moursi et al. (2008); Daniels, Adair, Popkin, and Truong (2009), Kennedy, Pedro, Seghieri, Nantel, and Brouwer (2007) and Steyn, Nel, Nantel, Kennedy, and Labadarios (2006). 6 Intensive breastfeeding: majority of the nutrition intake (80-100 percent) comes from breastmilk. 9 Figure 3.1. Distribution of children's dietary diversity score Source: own calculation from DHS-2016. Next, we exclude those children who did not consume any solid or semi-solid foods in the last 24 hours and assess the diet content. Figure 3.2 shows the percent of children consuming from the seven food groups. Pastoralist and agro-pastoralism are important livelihoods in the region and therefore it is not surprising that food consumption among young children is concentrated on dairy products. Grains, roots, or tubers were consumed by 68 percent of the children and legumes or nuts by 16 percent. Consumption from other food groups was less common, the share of children consuming from these food groups ranging between 2 and 6 percent. 10 Figure 3.2. Percent of children consuming from different food groups Source: own calculation from DHS-2016. Note: sample restricted to children who did consume foods in the last 24 hours. We then wondered how dietary diversity changes as children age. To assess this, we regressed dietary diversity score (number of food groups consumed by the child) on child's age using local polynomial regression approach. Figure 3.3 shows how dietary diversity increases relatively sharply between 6-9 months suggesting that the introduction of complementary foods is delayed (see above). Alternatively, caregivers do not give certain food types (e.g. fruits, vegetables, or animal-source foods) to very young children due to misperceptions that children cannot digest these (for evidence on this across Ethiopia, see Alive & Thrive, 2010; Hadley, Lindstrom, Belachew, & Tessema, 2008; USAID, 2011). After this, the mean dietary diversity score remains stable and the average child in the 18-23-month age range consumes from one to two food groups. 11 Figure 3.3. Relationship between child dietary diversity and child's age Note: Local polynomial regression. Shaded areas represent 95% confidence intervals. Source: own calculation from DHS-2016. There is an emerging consensus among researchers and practitioners that income growth alone will not address poor dietary quality (Ruel, Alderman, Maternal, & Group, 2013). To illustrate this, we regressed child dietary diversity on an aggregated wealth measure based on household's ownership of different durable assets. Figure 3.4 shows how children originating from wealthier households do consume a more diverse diet but even in the richest households, the average child consumes only from two food groups. 12 Figure 3.4. Relationship between child dietary diversity and household wealth Note: Local polynomial regression. Shaded areas represent 95 percent confidence intervals. The vertical dashed line represents 90th percentile of the wealth index distribution; 90 percent of the (child) sample are located on the left side of this line. Source: own calculation from DHS-2016. 3.2. Household diets7 Dietary diversity at the household level is measured as the mean number of food groups consumed. The indicator reports the number of food groups consumed in the seven days preceding the interview. Following the FANTA guidelines (Swindale & Bilinsky, 2006), we group households' responses to these questions into 12 food groups: 1. cereals; 2. root and tubers; 3. vegetables; 4. fruits; 7 Data on women's or mother's diets are not available to us for Afar. The DHS does not collect data on women's dietary patterns. Also, none of the IFPRI or CSA surveys administered in Afar collected these types of data. It is for this reason we report on household dietary diversity in this section. 13 5. meat, poultry, and offal; 6. eggs; 7. fish and seafood; 8. pulses, 9. legumes and nuts; 10. milk and milk products; 11. oil and fats; 12. sugar and honey; 13. miscellaneous foods. We then assign a value of one if the household consumed from the food group and zero if they did not. Totaling these values gives a household dietary diversity score (HDDS) that ranges between zero and 12. The average household in the PSNP-2018 sample consumed from 4.6 food groups (out of 12). Figure 3.5 shows the full distribution of household's dietary diversity score in the PSNP woredas in June/July 2018. About 7 percent of the households consumed only from one food group, 12 percent from two food groups and 17 percent from three food groups. Nearly 16 percent consumed from four food groups, 15 percent from five and 13 percent from six. About 11 percent of the households in Afar consumed from more than six food groups. Figure 3.5. Distribution of household dietary diversity score in the PSNP areas of Afar Source: own calculation from PSNP-2018. 14 Figure 3.6 shows the percent of households consuming from the 12 food groups. More than 80 percent of the households consumed cereals, 18 percent roots or tubers. Vegetables were consumed by 53 percent and fruits by 20 percent of the households. From animal-source foods, dairy products were consumed by 58 percent of the households, meat-by products by 38 percent, eggs by 4 percent, and fish by 2 percent of the households. Finally, pulses were consumed by 30 percent of the households. Figure 3.6. Percent of households consuming from different food groups in the PSNP areas of Afar Source: own calculation from PSNP-2018. 4. Food production from a nutrition perspective In this section, we examine the agricultural production in the region from a nutrition perspective. Using CSA's agricultural sample survey data, we disaggregate the production into the seven-food groups used to measure children's dietary diversity in section 3.1. To capture the foods that are not categorized in the seven-food groups (e.g. red peppers, garlic, coffee, sugarcane, and honey), we add an additional food group and called it ‘other foods.’ We then convert the amounts produced to calories using the Ethiopian food composition tables (EPHI, undated) and FAO reject estimates which correspond to weight estimations of edible portions of foods produced to 15 calculate energy supplied. The total energy are then divided by the total population for that specific year reported in CSA (2013). Finally, the energy produced is expressed on a per day basis by dividing by 365 to enable comparison with daily requirements. 8 Table 4.1 shows the daily total energy production per capita per these food groups. We begin by noting that crop production in Afar is low and rainfall patterns relatively erratic. As a result, agricultural output fluctuates considerably across years. First, the total calorie production declined by 6 percent between 2012 and 2015. This decrease was largely driven by the production of grains, roots, and tubers. In this food group, the amount of calories produced decreased by 16 percent. Second, we see that the agricultural production is concentrated on the production of grains, roots, and tubers that accounted for 64 percent of all calories produced in the region in 2015. 9 Dairy products account for 35 percent of the total calorie production in the region. Strikingly, the remaining six food groups account only for about 1 percent of the total calories produced. Finally, we note that the estimated total amount of energy and nutrients produced in Afar is not sufficient to feed the region's population; see Table 4.2. Consequently, the region must be a net-food importer. Table 4.1. Total energy production per capita per day, 2012-2015, by food categories 2012 2013 2014 2015 Grains, roots, and tubers 510 387 374 427 Legumes and nuts 0.0 0.0 0.4 0.0 Dairy products 195 224 243 233 Poultry, fish, meat 7.2 6.9 2.1 5.8 Eggs 0.2 0.1 0.3 0.3 Vitamin A-rich fruits and vegetables 0.0 0.0 0.0 0.0 Other fruits and vegetables 0.1 0.1 0.0 0.0 Other foods 0.3 0.1 0.1 0.0 Total 712 618 620 666 8 The implicit assumption here is that the availability of energy (and nutrients) is constant across seasons. This is unlikely to be the case: agricultural production in Ethiopia is largely rain-fed (Taffesse, Dorosh, & Gemessa, 2012), and therefore highly seasonal. Hirvonen, Taffesse, and Worku (2016) show how household energy intakes vary across seasons. 9 It is instructive to compare this production share to international dietary recommendations. For example, the recently developed diet recommendation by the EAT-Lancet commission recommends that only one-third of the total daily calories come from grains, roots, and tubers (Willett et al., 2019). 16 Source: Own calculation from CSA agricultural sample surveys. Amounts are kilocalories. Table 4.2 Prevalence (%) of energy and nutrient gaps relative to population-adjusted estimated average requirements in 2015 Afar Energy (kcal) 99.8 Protein (g) 96.9 Vitamin A (µg RAE) 16.5 Ascorbic acid (mg) 42.6 Iron (mg) 11.4 Zinc (mg) 39.8 Calcium (mg) 92.2 Thiamine (mg) 96.1 Niacin (mg) 91.2 Riboflavin (mg) 95.5 Source: Baye et al. (2019) Several limitations need to be considered when interpreting the results presented in this section. First, these estimates ignore international trade in agricultural products. However, exports form only 2 percent of the total domestic production. Meanwhile, the FAO food balance sheet data indicates that less than 4 percent of the domestic food supply in Ethiopia comes from imports, out of which nearly all are cereals (Baye et al., 2019). Consequently, the estimates are likely to be slightly underestimated, especially for the energy produced in the country. Second, there are several factors which limit the extent to which the produced calories and nutrients are actually available for human consumption. These factors include post-harvest losses, food waste, seasonality of food production, food prices, food preparation, and intra-household distribution, which all affect what is actually consumed from what is produced. That said, the purpose of this exercise is to highlight what could be achieved with more efficient post-harvest handling, better market integration, storage, and processing infrastructures along with improved consumer behavior. 5. Market availability of nutritious foods In this section, we assess the availability of nutritious foods in rural markets. As shown by Sibhatu and Qaim (2017), nutritious foods, such as fruits, vegetables, meat and eggs, are largely sourced from markets in rural Ethiopia. The exception is dairy for which the markets are missing or poorly functioning (Hoddinott, Headey, & Dereje, 2015). 17 The PSNP survey teams visited 36 communities in rural Afar. In each community, the team visited the local food market to assess food availability and prices. A novel feature of this market survey was that the enumerators recorded whether the food item is available in the market.10 We group these market data using the 10-food group categorization often used to analyze women's dietary diversity (FAO & FHI 360, 2016), with the exception that we do not include the starchy staple food group in the analysis.11 Table 5.1 shows the market availability in the beans and peas food group. In this food group, lentils are the product that is most often available (72 percent of the markets). Horse beans and chickpeas are available in about half of the markets where PSNP operates. Table 5.1. Market availability of food items in the Beans and peas food group, by survey Beans and peas (7 items in both surveys) PSNP (%) Bean, white 17 Bean, brown 0 Cowpea 0 Horse bean 47 Chick pea 56 Lentils 72 Green bean 8 Source: Own calculation from PSNP-2018 (June-July 2018) and FtF (September-October 2018) surveys. The market availability in the nuts and seeds food group is summarized in Table 5.2. Groundnut and fenugreek are available in more than half of the markets. The availability of other foods in this food group is limited. 10 Standard market surveys do not typically collect this information. As a result, the end-user of the data cannot be sure whether the price observation is missing because the item did not exist in the market or because of a data entry or other mistake. 11 It is tempting to compare these market availability findings to the production figures presented in Section 4. There are several factors at play why the market availability and agricultural output (measured in calories) figures seem out of sync. First, the market availability analysis does not consider price. Indeed, foods may be available in the markets but because of limited supply the prices are high, thereby excluding the poorest consumers from purchasing these items. We assess affordability in the next section. Second, for some of these items the calorie density (kcal/kg) is low. This is particularly the case for green leafy vegetables such as kale and spinach. Third, production of many fruits and vegetables is largely seasonal, which means that the total annual production may be low while the production (and hence) availability in one particular season is high. Fourth, some of the foods sold at the markets may come from other regions. 18 Table 5.2. Market availability of food items in the Nuts and seeds food group, by survey Nuts and seeds (5 items in both surveys) PSNP (%) Groundnut 53 Groundnut flour 0 Fenugreek 64 Vetch 17 Sesame 14 Source: Own calculation from PSNP-2018 (June-July 2018) survey. Table 5.3 shows the market availability for the items in the dairy food group. Fresh milk is available in less than 30 percent of the markets, yoghurt in 25 percent and cheese only in 6 percent of the markets. Powdered milk is available in 36 percent of the markets. Table 5.3. Market availability of food items in the Dairy food group, by survey Dairy (4 items in PSNP, 5 in FtF) PSNP (%) Fresh milk 28 Powdered milk 36 Yoghurt 25 Cheese 6 Source: Own calculation from PSNP-2018 (June-July 2018) survey. Beef was available in less than 20 percent of the markets (Table 5.4). Goat and mutton in 69 and 22 percent of the markets, respectively. Chicken was available in 22 percent of the market while camel meat only in 6 percent of the markets. 19 Table 5.4. Market availability of food items in the Flesh foods food group, by survey Flesh foods (8 items in PSNP, 9 in FtF) PSNP (%) Dried fish 0 Fresh fish 6 Beef 19 Goat 69 Mutton 22 Chicken 22 Camel meat 6 Source: Own calculation from PSNP-2018 (June-July 2018) survey. Eggs are available in more than 80 percent of the markets (Table 5.5) but the availability of Vitamin A-rich dark green leafy vegetables varies (Table 5.6). Ethiopian kale (gommen) is available in 47 percent of the PSNP markets while 56 percent had spinach. Table 5.5. Market availability of food items in the Egg food group, by survey Eggs (1 item in both surveys) PSNP (%) Eggs 83 Source: Own calculation from PSNP-2018 (June-July 2018) survey. Table 5.6. Market availability of food items in the Vitamin A-rich dark green leafy vegetable food group, by survey Vitamin A-rich dark green leafy vegetables (2 items in both surveys) PSNP (%) Ethiopian Kale 47 Spinach 56 Source: Own calculation from PSNP-2018 (June-July 2018) survey. Table 5.7 shows the availability of foods under the other Vitamin Arich vegetable and fruit group. At the time the enumerators in the PSNP survey visited the markets, carrot and mangos were found in more than 80 percent of the markets in Afar and papayas in 47 percent of the markets. Table 5.7. Market availability of food items in the Other vitamin A-rich vegetables and fruit food group, by survey Other Vitamin A-rich vegetables and fruits (3 items in PSNP, 4 in FtF) PSNP (%) Carrot 72 Mango 72 Papaya 47 Source: Own calculation from PSNP-2018 (June-July 2018) survey. 20 Other types of vegetables are widely available (Table 5.8). Almost all markets had onions, and tomatoes. Green peppers and lettuce were found in more than 50 percent of the markets. Mushrooms and cauliflower were hardly available. Table 5.8. Market availability of food items in the Other vegetable food group, by survey Other vegetables (6 items in both surveys) PSNP (%) Onion 94 Tomato 100 Mushroom 3 Lettuce 56 Cauliflower 0 Green pepper 67 Source: Own calculation from PSNP-2018 (June-July 2018) survey. Other fruits (that are not rich in Vitamin A) are not widely available in Afar (Table 5.9). The exceptions are bananas that were found in more than 80 percent of the markets. At the time of the survey, none of the markets had oranges or cactus fruits. Table 5.9. Market availability of food items in the Other fruit food group, by survey Other fruits (3 items in PSNP, 6 in FtF) PSNP (%) Banana 83 Orange 0 Cactus fruit 0 Source: Own calculation from PSNP-2018 (June-July 2018) survey. 6. Affordability of nutritious foods In this section, we assess the affordability of nutritious foods in Afar. We use the seven-food group categorization used to assess children's diets in Section 2.1, except we omit the first food group: grains, roots, and tubers. We define affordability as the share of total income needed to consume the recommended daily amount of the food group.12 Ethiopia has not yet developed itsown nutritional guidelines or eating recommendations. Therefore, we have to use international nutritional guidelines to determine the recommended intake for each food group. Here, we use the recently developed recommendations by the EAT- 12 These calculations do not account for refuse factors and therefore these affordability estimates should be considered as lower bound. 21 Lancet Commission on Food, Planet and Health (Willett et al., 2019). The EAT diet recommendations attempt to maximize health benefits while minimizing the diet's negative impact on environment. Hence, the diet recommendation puts more emphasis on plant-based foods (legumes, nuts, vegetables, and fruits), and less on meat and eggs. 13 Table 6.1 maps these recommendations to the six food groups. Table 6.1. EAT-Lancet Commission's intake recommendations, by food group Food group grams per day Legumes and nuts 125 Dairy products 250 Poultry, fish, and meat 71 Eggs 13 Vitamin A- rich foods and fruits 200 Other fruits and vegetables 300 Source: Willet et al (2019). Of note is that nutritional needs vary by age, sex, and physical activity levels. The EAT-Lancet Commission's dietary recommendations are for healthy individuals 2 years or older. Infants and young children (less than 2 years of age) are growing rapidly and therefore have very different requirements than others (Willett et al., 2019). The purpose of the affordability analysis carried out here is not to calculate the cost of an optimal diet for children or women. Rather, we aim to provide a sense of the price of these foods relative to the household income levels in the region. For income, we rely on CSA’s Household Consumption Expenditure (HCE) survey report for the region based on data collected between July 2015 and June 2016 (CSA, 2018a). The HCE survey does not record incomes, and, therefore, we proxy household incomes using CSA’s estimates of household consumption-expenditures. The mean annual per capita consumption expenditure in 13 As noted by Willett et al (2019), the recommended amount of animal-source foods may be sub-optimal for many sub-Saharan African countries. More specifically, the authors write (p.10) that "[b]ecause many regions, such as sub-Saharan Africa, still face severe burdens of undernutrition and malnutrition, and growing children often do not obtain adequate quantities of nutrients from plant-source foods alone, the role of animal-source foods should be examined carefully". 22 the region was estimated as 8,992 birr (CSA, 2018a). Table 6.2 shows the mean annual per capita consumption expenditure estimates for each quintile. Table 6.2. Annual household per capita expenditure in 2015/16 Statistic Amount (birr) Mean 8,992 Poorest quintile 4,498 2nd quintile 6,201 3rd quintile 8,887 4th quintile 12,517 Richest quintile 24,328 Source: CSA (2018a). We calculated the mean price in the region for food items using CSA’s monthly retail price data for the same period (2015/16). To address the seasonality in food availability and prices, we selected the cheapest item in each food group in each month. Table 6.3 below provides the list of the cheapest food items in each category. It is important to note that the set of foods picked as the cheapest one in Table 6.3 may not be widely available or traditionally consumed in the region. However, the purpose of this exercise is to calculate the minimum cost of the dietary recommendation. Any departure from these items means that the estimated cost will increase further. Table 6.3. Cheapest food item in each food group Year Month Legumes and nuts Dairy products Poultry, fish, and meat Eggs Vitamin A- rich fruits and vegs Other fruits and vegetables Average Haricot beans Cow milk Camel meat Egg Pumpkin Cabbage 2015 July Vetch Cow milk Camel meat Egg Pumpkin Onion August Lima beans Cow milk Camel meat Egg Pumpkin Onion September Lima beans Cow milk Beef Egg Pumpkin Cabbage October Lima beans Cow milk Beef Egg Pumpkin Cabbage November Vetch Cow milk Beef Egg Pumpkin Cabbage December Lima beans Cow milk Camel meat Egg Pumpkin Cabbage 2016 January Haricot beans Cow milk Camel meat Egg Pumpkin Onion February Haricot beans Cow milk Beef Egg Spinach Onion March Haricot beans Cow milk Camel meat Egg Pumpkin Onion April Haricot beans Camel milk Camel meat Egg Pumpkin Onion May Vetch Cow milk Camel meat Egg Kale cabbage June Haricot beans Cow milk Camel meat Egg Pumpkin Cabbage Note: Cow milk refers to unpasteurized milk. Kale is Ethiopian Kale (gommen). Source: Own calculation from CSA price data. 23 We then used the prices of the above food items to calculate the total annual cost of consuming the recommended serving reported in Table 6.1. To get the share of total income needed to meet the recommended intake, we divided the total annual cost of the recommended intake by the annual per capita consumption expenditure value divided by 365 (to get the daily cost per person). Table 6.4 provides the results. 14 Following the EAT diet would mean that the average resident in Afar would spend 6 percent of their income on legumes and nuts, 18 percent on dairy, 34 percent on meat, 5 percent on eggs and 6 percent on Vitamin A-rich fruits and vegetables, and 12 percent of other fruits and vegetables. These percentages are much higher for the poorest households for which many of these recommendations are out of reach. Therefore, the poorest households may require special attention when nutrition-sensitive agriculture and other policies are planned.15 Despite the moderate emphasis on meat products in the EAT diet, consuming the recommended intake (71 grams) would be quite expensive in Afar. For the average resident in the region, this would mean spending 33 percent of their budget just on meat products. For the poorest households, a daily ration of 71 grams of meat would take 68 percent of their total budget. The EAT-Lancet commission recommends consuming 13 grams of eggs every day. Considering that the local egg weighs about 35 grams, this means consuming two to three eggs per week. For the average resident in Afar, this would mean allocating 5 percent of the budget on eggs. For the poorest, the corresponding share would be 10 percent. A large part of the daily calories in the EAT diet comes from the consumption of fruits and vegetables. Here, we have divided these into items that are rich in Vitamin A and other items. Consuming 200 grams of fruits and vegetables that are rich in Vitamin A would mean allocating 6 percent of the budget on these items at the mean-income level. For the poorest, the corresponding share is 11 percent. Following the recommended intake for other fruits and 14 It is worth noting here that as per the Engel's law, the average Ethiopian allocates a large fraction of their budget on food. Therefore the shares allocated to food are high, irrespective of which diet recommendation we use. 15 It is important to note here that nutrition provisions have recently been integrated into Ethiopia's productive safety net program (PSNP) that covers 8 million poor rural people (Bossuyt, 2017). 24 vegetables, would take 12 percent of the total budget at the mean-income level, and 24 percent at income level of the poorest households. Table 6.4. Percent of household income needed to meet the recommended intake, by income quintile Mean income Poorest quintile 2nd quintile 3rd quintile 4th quintile Richest quintile Legumes and nuts 6 11 8 6 4 2 Dairy products 18 35 26 18 13 7 Poultry, fish, and meat 34 68 49 34 24 13 Eggs 5 10 8 5 4 2 Vitamin A-rich fruit and vegs 6 11 8 6 4 2 Other fruits and vegetables 12 24 17 12 9 4 Total 81 159 116 81 58 30 Finally, pumpkin and cabbage were identified as the cheapest items in their food groups (Table 6.3). These foods are not widely consumed in Ethiopia, perhaps because they can only be bought in large quantities (in terms of weight) and in absence of preservation technologies, are not actually affordable. This prompts us to explore the sensitivity of these estimates considering the second cheapest item in the fruit and vegetable food groups (see Appendix A). These results are presented in Table 6.5. As expected, the percent of income needed to meet the recommended intake somewhat increases. For the average resident, following the EAT recommendation for Vitamin A-rich fruit and vegetables would cost 12 percent of the total budget. The corresponding share among the poorest households is 23 percent. As for other fruits and vegetables, 16 percent is needed at the mean-income level and 32 percent at the income level of the poorest household. Table 6.5. Percent of household income needed to meet the recommended intake for fruits and vegetables, second cheapest items, by income quintile Mean income Poorest quintile 2nd quintile 3rd quintile 4th quintile Richest quintile Vitamin A-rich foods and fruits 12 23 17 12 8 4 Other fruits and vegetables 17 35 25 17 12 6 25 7. Conclusions This report has analyzed the availability, access, and consumption of nutritious foods in the Afar region of Ethiopia. In this section, we summarize the findings. Children's diets in the region are extremely monotonous. Only 2 percent of the children 6-23 months meet the WHO recommended dietary diversity (four food groups out of seven). Further analysis suggest that one reason for poor dietary diversity among infants and young children is delayed initiation of complementary foods. Other possible explanations include misperceptions around when to introduce certain foods to young children as well as poor affordability of certain foods, such as ASFs. Nearly 39 percent of children would need to add one or two additional food groups to their diet to meet the four-food group recommendation. Should this happen, the share of children achieving the minimum dietary diversity would increase to 41 percent. With this in mind, the following passages discuss the non-starchy staple food groups and their potential of reaching the most households in the region to improve dietary diversity. Before doing so, we highlight that the Afar region is characterized by extremely challenging agro-climatic conditions. The lack of rainfall and absence of irrigation, in particular, limit the development of crop agriculture and as a result the region is a net-food importer (see section 4). This needs to be taken into account when nutrition-sensitive agriculture strategies are drafted. 7.1. Legumes and nuts Legumes and nuts are not widely consumed among children in Afar. Meanwhile, they are widely available in the rural markets and their production is at a relatively reasonable level, supplying nearly 380 calories per person per day. They also are affordable. Therefore, legumes and nuts have a high potential for improving dietary diversity in Afar. 7.2. Dairy products Dairy products are consumed by more than half of the children and their production level is relatively high, supplying about 230 calories per day per person. Meanwhile, their market availability and affordability is poor. We take this to suggest that milk markets are largely missing in Afar, and that consumption is largely sourced from own production or through barter with neighbors. While dairy is an important source of many nutrients and highly beneficial input for child growth (Headey, Hirvonen, & Hoddinott, 2018), their commercial value chains are 26 extremely complex, requiring refrigeration or industrial processing (e.g. powder or ultra-high temperature processing) to maintain food safety. Given all this, increasing dairy consumption further in Afar, especially in rural areas, would require considerable investments. 7.3. Flesh foods Few children consume meat products in Afar. Their production is relatively low, supplying only about 6 calories per day per person. Furthermore, while their market availability is poor, meat products are not affordable for most households. All things considered, improving the availability and affordability of meat products is challenging in the Afar region and will require innovative solutions. Dried shredded or powdered meat products hold promise as they simultaneously solve several issues regarding perishability, affordability (can be sold in small portions), and acceptability among young children. 7.4. Eggs Egg consumption is not widespread among young children in Afar. Also their production levels are relatively low, supplying less than 0.5 calories per person per day. However, their market availability is good with most rural markets stocking eggs. They are also relatively affordable, and have high nutritional content and recent evidence suggests eggs have a high potential for reducing stunting rates in poor countries (Headey et al., 2018; Iannotti, Lutter, Bunn, & Stewart, 2014; Iannotti et al., 2017). Therefore, eggs could have a high potential for improving dietary diversity in Afar. 7.5. Vitamin - rich fruits and vegetables Vitamin A-rich fruits and vegetables are not consumed widely by children in the region. Their market availability is also good and they are affordable. Therefore, Vitamin A-rich fruits and vegetables could have a high potential for improving dietary diversity in Afar. However, high perishability and the seasonal availability of these products should be taken into account. 7.6. Other fruits and vegetables The consumption of other types of fruits and vegetables is even less widespread among children. Meanwhile, their market availability in the region is good with most markets stocking onions, tomatoes, and bananas. They are also reasonably affordable. Therefore, other fruits and vegetables may also have a potential for improving dietary diversity in Afar. However, high perishability and the seasonal availability of these products should be taken into account. 27 7.7. Limitations This study has limitations. First, data limitations prevent us from analyzing women's dietary diversity in the region. Second, although we use a geographically widespread PSNP survey to assess market availability of nutritious foods, in the absence of regionally representative data, we cannot be sure whether the statistics reported here represent the region as a whole. Third, our estimates on crop production levels is limited to the production during the main rainy seasons. Significant crop production outside the rainy seasons in the region (e.g. using irrigation) would mean that production levels are under-estimated. 28 References Abebe, Z., Haki, G. D., Schweigert, F. J., Henkel, I. M., & Baye, K. (2018). Low breastmilk vitamin A concentration is prevalent in rural Ethiopia. European Journal of Clinical Nutrition. Alive & Thrive. (2010). IYCF practices, beliefs and influences in Tigray Region, Ethiopia., Addis Ababa, Ethiopia: Alive & Thrive. Arimond, M., & Ruel, M. T. (2004). Dietary diversity is associated with child nutritional status: evidence from 11 demographic and health surveys. The Journal of Nutrition, 134(10), 2579-2585. Bachewe, F., Hirvonen, K., Minten, B., & Yimer, F. (2017). The rising costs of nutritious diets in Ethiopia ESSP Research Note, 67, Addis Ababa: IFPRI. Baye, K., Hirvonen, K., Dereje, M., & Remans, R. (2019). Energy and nutrient production in Ethiopia, 2011-2015: implications to supporting healthy diets and food systems. PloS one. CSA. (2013). Population Projection of Ethiopia for All Regions at Wereda Level from 2014 – 2017, Addis Ababa, Ethiopia: Central Statistical Agency (CSA) of Ethiopia. CSA. (2018a). The 2015/16 Ethiopian Household Consumption – Expenditure (HCE) Survey: Country Level Statistical Report, Addis Ababa: Central Statistical Agency (CSA) of the Federal Democratic Republic of Ethiopia CSA. (2018b). The 2015/16 Ethiopian Household Consumption Expenditure (HCE) Survey: Results for Afar region. Addis Ababa, Ethiopia: Central Statistical Agency (CSA) of Ethiopia. CSA, & ICF. (2016). Ethiopia Demographic and Health Survey 2016, Addis Ababa, Ethiopia, and Rockville, Maryland, USA: Central Statistical Agency (CSA) of Ethiopia and ICF. Daniels, M., Adair, L., Popkin, B., & Truong, Y. (2009). Dietary diversity scores can be improved through the use of portion requirements: an analysis in young Filipino children. European Journal of Clinical Nutrition, 63(2), 199-208. Dehghan, M., Mente, A., Zhang, X., Swaminathan, S., Li, W., Mohan, V., . . . Rosengren, A. (2017). Associations of fats and carbohydrate intake with cardiovascular disease and mortality in 18 countries from five continents (PURE): a prospective cohort study. The Lancet, 390(10107), 2050-2062. Demissie, T., Ali, A., Mekonen, Y., Haider, J., & Umeta, M. (2010). Magnitude and distribution of vitamin A deficiency in Ethiopia. Food and Nutrition Bulletin, 31(2), 234-241. EPHI. (2016). Ethiopian National Micronutrient Survey Report, Addis Ababa: Ethiopian Public Health Institute (EPHI). EPHI. (undated). Expanded Food Composition Table for Use in Ethiopia. Addis Ababa Ethiopian Public Health Institute (EPHI). FAO. (1972). Technical conversion factors for agricultural commodities. Rome: Statistics Division of the Food and Agriculture Organization (FAO) of the United Nations 29 FAO, & FHI 360. (2016). Minimum Dietary Diversity for Women: A Guide to Measurement. Rome: Food and Agriculture Organization (FAO) of the United Nations and USAID’s Food and Nutrition Technical Assistance III Project (FANTA), managed by FHI 360. Forouzanfar, M. H., Alexander, L., Anderson, H. R., Bachman, V. F., Biryukov, S., Brauer, M., . . . Murray, C. J. (2015). Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks in 188 countries, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. The Lancet, 386(10010), 2287-2323. Gebru, M., Remans, R., Brouwer, I., Baye, K., Melesse, M., Covic, N., . . . Hirvonen, K. (2018). Food systems for healthier diets in Ethiopia: Toward a research agenda. GFDRE. (2016a). National Nutrition Programme 2016-2020. Addis Ababa: Government of the Federal Democratic Republic of Ethiopia (GFDRE). GFDRE. (2016b). Seqota Declaration Implementation Plan (2016 – 2030): Summary Programme Approach Document, Addis Ababa: Government of the Federal Democratic Republic of Ethiopia (GFDRE). Hadley, C., Lindstrom, D., Belachew, T., & Tessema, F. (2008). Ethiopian adolescents' attitudes and expectations deviate from current infant and young child feeding recommendations. Journal of Adolescent Health, 43(3), 253-259. Hall, J. N., Moore, S., Harper, S. B., & Lynch, J. W. (2009). Global Variability in Fruit and Vegetable Consumption. American Journal of Preventive Medicine, 36(5), 402-409.e405. doi:https://doi.org/10.1016/j.amepre.2009.01.029 Hassen Worku, I., Dereje, M., Minten, B., & Hirvonen, K. (2017). Diet transformation in Africa: the case of Ethiopia. Agricultural Economics, 48(supplement), 73–86. Headey, D., Nisrane, F. B., Worku, I., Dereje, M., & Taffesse, A. S. (2012). Urban wage behavior and food price inflation: The case of ethiopia. Ethiopia Strategy Support Programme II Working Paper, 41. Headey, D. D., Hirvonen, K., & Hoddinott, J. F. (2018). Animal sourced foods and child stunting. American Journal of Agricultural Economics, 100(5), 1302–1319. Hirvonen, K. (2016). Rural-urban differences in children's dietary diversity in Ethiopia: a Poisson decomposition analysis. Economics Letters, 147, 12-15. Hirvonen, K., Headey, D., Golan, J., & Hoddinott, J. (2019). Changes in child undernutrition rates in Ethiopia 2000-2016. In C. Cramer, F. Cheru, & A. Oqubay (Eds.), The Oxford Handbook of the Ethiopian Economy. Oxford: Oxford University Press. Hirvonen, K., Taffesse, A. S., & Worku, I. (2016). Seasonality and household diets in Ethiopia. Public Health Nutrition, 19(10), 1723-1730. Hoddinott, J., Headey, D., & Dereje, M. (2015). Cows, missing milk markets and nutrition in rural Ethiopia. Journal of Development Studies, 51(8), 958–975. Iannotti, L. L., Lutter, C. K., Bunn, D. A., & Stewart, C. P. (2014). Eggs: the uncracked potential for improving maternal and young child nutrition among the world's poor. Nutrition Reviews, 72(6), 355-368. https://doi.org/10.1016/j.amepre.2009.01.029 30 Iannotti, L. L., Lutter, C. K., Stewart, C. P., Riofrío, C. A. G., Malo, C., Reinhart, G., . . . Cox, K. (2017). Eggs in Early Complementary Feeding and Child Growth: A Randomized Controlled Trial. Pediatrics, e20163459. Jones, A. D., Ickes, S. B., Smith, L. E., Mbuya, M. N., Chasekwa, B., Heidkamp, R. A., . . . Stoltzfus, R. J. (2014). World Health Organization infant and young child feeding indicators and their associations with child anthropometry: a synthesis of recent findings. Maternal & child nutrition, 10(1), 1-17. Kennedy, G. L., Pedro, M. R., Seghieri, C., Nantel, G., & Brouwer, I. (2007). Dietary diversity score is a useful indicator of micronutrient intake in non-breast-feeding Filipino children. The Journal of Nutrition, 137(2), 472-477. Menon, P., & Frongillo, E. A. (2018). Can integrated interventions create the conditions that support caregiving for better child growth? The Lancet Global Health, 6(3), e236-e237. Misganaw, A., Haregu, T. N., Deribe, K., Tessema, G. A., Deribew, A., Melaku, Y. A., . . . Dessalegn, M. (2017). National mortality burden due to communicable, non- communicable, and other diseases in Ethiopia, 1990–2015: findings from the Global Burden of Disease Study 2015. Population health metrics, 15(1), 29. Moursi, M. M., Arimond, M., Dewey, K. G., Trèche, S., Ruel, M. T., & Delpeuch, F. (2008). Dietary diversity is a good predictor of the micronutrient density of the diet of 6-to 23- month-old children in Madagascar. The Journal of Nutrition, 138(12), 2448-2453. Ruel, M. T. (2003). Operationalizing dietary diversity: a review of measurement issues and research priorities. The Journal of Nutrition, 133(11), 3911S-3926S. Ruel, M. T., Alderman, H., Maternal, & Group, C. N. S. (2013). Nutrition-sensitive interventions and programmes: how can they help to accelerate progress in improving maternal and child nutrition? The Lancet, 382(9891), 536-551. Scott, P. (2017). Global panel on agriculture and food systems for nutrition: food systems and diets: facing the challenges of the 21st century. Food Security: The Science, Sociology and Economics of Food Production and Access to Food, 9(3), 653-654. Sibhatu, K. T., & Qaim, M. (2017). Rural food security, subsistence agriculture, and seasonality. PloS one, 12(10), e0186406. Steyn, N., Nel, J., Nantel, G., Kennedy, G., & Labadarios, D. (2006). Food variety and dietary diversity scores in children: are they good indicators of dietary adequacy? Public Health Nutrition, 9(05), 644-650. Swindale, A., & Bilinsky, P. (2006). Household dietary diversity score (HDDS) for measurement of household food access: indicator guide. Washington, DC: Food and Nutrition Technical Assistance Project, Academy for Educational Development. Taffesse, A. S., Dorosh, P., & Gemessa, S. A. (2012). Crop Production in Ethiopia. In P. Dorosh & S. Rashid (Eds.), Food and Agriculture in Ethiopia. Philadelphia: University of Pennsylvania Press. USAID. (2011). Integration of nutrition education into the Ethiopia Urban Gardens Program: Results of recipe trials and focus group discussions. Infant & Young Child Nutrition Project, Washington DC.: United States Agency for International Development (USAID). 31 WHO. (2008). Indicators for assessing infant and young child feeding practices: part 1: definitions: conclusions of a consensus meeting held 6-8 November 2007 in Washington DC, USA. Geneva: World Health Organization (WHO). WHO. (2010). Monitoring the building blocks of health systems: a handbook of indicators and their measurement strategies. Geneva: World Health Organization (WHO). Willett, W., Rockström, J., Loken, B., Springmann, M., Lang, T., Vermeulen, S., . . . Murray, C. J. L. (2019). Food in the Anthropocene: the EAT–Lancet Commission on healthy diets from sustainable food systems. The Lancet, Published Online January 16, 2019. http://dx.doi.org/10.1016/S0140-6736(18)31788-4, 1-47. Zerfu, T. A., Pinto, E., & Baye, K. (2018). Consumption of dairy, fruits and dark green leafy vegetables is associated with lower risk of adverse pregnancy outcomes (APO): a prospective cohort study in rural Ethiopia. European Journal of Clinical Nutrition, 8(52). http://dx.doi.org/10.1016/S0140-6736(18)31788-4 32 Appendix A: Recalculating affordability considering the second cheapest food items in fruit and vegetable categories Table A1. Second Cheapest food item in each food group Year Month Vitamin A-rich fruits and vegs Other fruits and vegetables 2015 July Carrot Tomato August Spinach Tomato September Spinach Onion October Spinach Tomato November Ethiopian Kale Onion December Ethiopian Kale Onion 2016 January Ethiopian Kale Tomato February Ethiopian Kale Tomato March Spinach Tomato April Spinach Tomato May Spinach Onion June Ethiopian Kale Onion 1. Introduction 2. Description of the data sources used in this report 2.1. Demographic and Health Survey (DHS) data 2.2. PSNP evaluation survey 2.3. CSA agricultural sample survey data 2.4. CSA retail price survey data 3. Consumption of nutritious foods 3.1. Child diets 3.2. Household diets6F 4. Food production from a nutrition perspective 5. Market availability of nutritious foods 6. Affordability of nutritious foods 7. Conclusions 7.1. Legumes and nuts 7.2. Dairy products 7.3. Flesh foods 7.4. Eggs 7.5. Vitamin - rich fruits and vegetables 7.6. Other fruits and vegetables 7.7. Limitations References Appendix A: Recalculating affordability considering the second cheapest food items in fruit and vegetable categories