Global, regional and national trends

The 2016 Global Hunger Index (GHI) demonstrates substantial progress in terms of hunger reduction for the developing world. Whereas the 2000 GHI score for the developing world was 30.0, the 2016 GHI score is 21.3, showing a reduction of 29 per cent (Figure 2.1). Underlying this improvement are reductions since 2000 in each of the GHI indicators - the prevalence of undernourishment, child stunting (low height for age), child wasting (low weight for height), and child mortality. Yet, as this chapter reveals, there are great disparities in hunger at the regional, national, and subnational levels, and progress has been uneven. To succeed in the Sustainable Development Goal 2 (SDG2) of achieving Zero Hunger while leaving no one behind, it is essential to identify the regions, countries, and populations that are most vulnerable to hunger and undernutrition, and accelerate progress in these areas.
In terms of the indicators used in the GHI, the proportion of the population that is undernourished has declined from 18.5 per cent to 13.1 per cent since 2000. Among children under age five, 28.1 per cent - just more than one in four - are stunted, down from the 2000 rate of 37.8 per cent; and 8.4 per cent suffer from wasting, slightly down from 9.9 per cent in 2000. Finally, the under-five mortality rate dropped from 8.2 per cent in 2000 to 4.7 per cent in 2015.  Black et al. (2013) estimate that undernutrition causes almost half of all child deaths globally.
Large regional differences: In terms of the major regions of the developing world, Africa south of the Sahara and South Asia have the highest 2016 GHI scores, at 30.1 and 29.0, respectively (Figure 2.1). These scores reflect serious levels of hunger, and while the GHI scores for these regions have declined over time, the current levels are still on the upper end of the serious category, closer to the alarming category (35.0-49.9) than to the moderate (10.0-19.9). The composition of the GHI scores varies between Africa south of the Sahara and South Asia. In South Asia, child undernutrition, as measured by child stunting and child wasting, is higher than in Africa south of the Sahara. On the other hand, the prevalence of undernourishment, reflecting overall calorie deficiency for the population, and the child mortality rate are higher in Africa south of the Sahara than in South Asia.
 The GHI scores for East and Southeast Asia, Near East and North Africa, Latin America and the Caribbean, and Eastern Europe and the Commonwealth of Independent States range between 7.8 and 12.8, and represent low or moderate levels of hunger. Yet disparities within each region are important to recognize, and certainly no conclusions can be drawn about a particular country based on the overall score of its larger geographical region. For example, Haiti has a 2016 GHI score of 36.9, which places it in the alarming category, despite being in Latin America and the Caribbean - the region of the developing world with the lowest GHI score. Also, the 2016 GHI score for East and Southeast Asia is 12.8, but this is strongly influenced by highly populous China, which has a low GHI score of just 7.7. Examination of the other countries in this grouping without China shows a GHI score of 19.9 - very near the threshold between the moderate and serious categories.
In terms of absolute change, Africa south of the Sahara has experienced the greatest improvement from the 2000 GHI to the 2016 GHI, with a reduction of 14.3 points. South Asia and East and Southeast Asia also reduced their GHI scores by sizable amounts since 2000, with reductions of 9.2 and 8.0 points, respectively. The Near East and North Africa, Latin America and the Caribbean, and Eastern Europe and the Commonwealth of Independent States reduced their GHI scores by 4.2 to 5.8 points. In terms of the percentage change relative to the 2000 GHI, Eastern Europe and the Commonwealth of Independent States, Latin America and the Caribbean, and East and Southeast Asia experienced declines of between 39 and 43 per cent. Africa south of the Sahara reduced its GH I score by 32 per cent, and the Near East and North Africa and South Asia reduced their scores by 26 per cent and 24 per cent, respectively.
The good news is that we do not see evidence of stagnation or stalled progress in terms of hunger reduction for any region. However, in the highest hunger regions, South Asia and Africa south of the Sahara, the rates of improvement must accelerate. If these regions were to reduce their hunger levels between 2016 and 2030 at the same pace of reduction they have experienced since 2000, they would still have GHI scores of roughly 20 to 22 points - at the low end of the serious category or on the border between moderate and serious - falling far short of the goal to reach Zero Hunger by 2030.
Country-level results: While we highlight many important findings in the following paragraphs, we invite the reader to use the facts that are contained herein to better understand the story of each country. Importantly, Appendix D shows the 1992, 2000, 2008, and 2016 GHI scores the same information and also gives each country's numerical ranking; the countries are ranked from best to worst performers based on their 2016 GHI scores. Appendix C shows the values of the GHI indicators - the prevalence of undernourishment, child wasting, child stunting, and child mortality - for each country, which form the basis of each country-level GHI score.
 From the 2000 GHI to the 2016 GHI, 22 countries made remarkable progress, reducing their GHI scores by 50.0 per cent or more (Figure 2.2). Seventy countries saw a considerable reduction in their scores, dropping by between 25.0 per cent and 49.9 per cent, and 22 countries decreased their GHI scores by less than 25.0 per cent. Despite this progress, 50 countries still suffer from serious or alarming levels of hunger.
Figure 2.3 includes the countries with serious or alarming hunger levels and shows both their 2016 GHI scores and the percentage reductions in their GHI scores since 2000. The Central African Republic and Chad, in the lower right-hand corner of the figure, are obvious areas of concern. These countries have the highest GHI scores in this year's report, coupled with relatively low percentage reductions in hunger since 2000. In the Central African Republic, violence and mass displacement resulting from a four-year-long civil war have taken a heavy toll on food production (FAO 2016a). Chad, which has also had a long history of civil war, has faced deteriorating food security, due in part to a recent influx of refugees and extreme weather events (FAO 2016b). The examples of these countries underscore that despite significant progress in reducing hunger globally, violent conflict, poor governance, and climate-related impacts on agriculture ensure that hunger continues to plague our planet and requires a transformative plan of action.
Namibia and Sri Lanka stand out for having the lowest percentage reductions in GHI scores since 2000. In the case of Sri Lanka, examination of its GHI indicator values reveals that while the prevalence of undernourishment, child stunting, and child mortality have declined moderately, child wasting has gone up and is the third-highest child wasting level in the report (Appendix C). The causes of child undernutrition in Sri Lanka are not well understood, but studies reveal multiple micronutrient deficiencies among children, with causes that include a combination of inadequate intake, as well as diseases that prevent proper nutrient absorption (Hettiarachchi and Liyanage 2010; Hettiarachchi and Liyanage 2012; Jayatissa et al. 2014).
In the case of Namibia, child stunting, child wasting, and child mortality have fallen, but the prevalence of undernourishment has risen since 2000, dragging down its overall score. Namibia is vulnerable to erratic rainfall, including frequent droughts and flooding, and has experienced drought for the last two to three years, putting downward pressure on Namibia's cereal and livestock production. Also, poor harvests within Namibia and in neighbouring countries have driven up food prices (WFP 2016b; FAO GIEWS 2016b).
Since 2000, Rwanda, Cambodia, and Myanmar, positioned at the top of Figure 2.3, have seen the largest percentage reductions in hunger of all the countries categorized as serious or alarming, with 2016 GHI scores down by just over 50 per cent relative to the 2000 scores in each country. Each of these countries has experienced civil war and political instability in recent decades, and the improvements may in part reflect increased stability. In the case of Myanmar, the improved score is driven by the falling prevalence of undernourishment, which has declined by nearly 75 per cent since 2000. It will be important to ensure that comparable gains in child nutrition are also realized. In Rwanda, child mortality and child wasting saw the biggest reductions, decreasing by approximately 75 per cent each; the prevalence of undernourishment fell by nearly half; and stunting only went down by 20 percent. In Cambodia, the child mortality rate went down the most, dropping by 73 per cent, while again child stunting decreased the least, falling by 34 per cent, which is still a strong improvement.
 Only one country - Kuwait - experienced an increase in its GHI score between 2000 and 2016. However, the increase is small in absolute terms, and Kuwait's hunger level is still categorized as low. Most importantly, Kuwait's 2016 score of 4.9 represents a dramatic improvement compared to its 1992 score, which rose to 26.0 in the wake of the Gulf War.
Seven countries still suffer from levels of hunger that are alarming. The majority of these are in Africa south of the Sahara: the Central African Republic, Chad, Madagascar, Sierra Leone, and Zambia. The exceptions are Haiti and the Republic of Yemen. While no countries had extremely alarming levels of hunger (GHI scores of 50 points or more) according to 2016 GHI scores, nine countries had extremely alarming hunger levels as recently as 2000: Afghanistan, Angola, the Central African Republic, Chad, Ethiopia, Niger, Rwanda, Sierra Leone, and Zambia. As in past years, GHI scores for several countries could not be included because data on undernourishment, and in some cases child stunting and child wasting, were not available. However, the hunger and/or undernutrition situations in many of these countries are cause for significant concern (Box 2.1) and in some cases could fall in the extremely alarming category.
In terms of the GHI components, Haiti, Zambia, and the Central African Republic have the highest proportion of undernourished people - between 47.7 per cent and 53.4 per cent of the population. Timor-Leste, Burundi, and Papua New Guinea have the highest prevalence of stunting, with approximately 50 per cent or more of children under age five suffering from stunting. South Sudan, Djibouti, and Sri Lanka have the highest prevalence of wasting, with between 21.4 per cent and 23.8 per cent of children under age five affected. Somalia, Chad, and Angola have the highest under-five mortality rates, ranging between 13.7 per cent and 15.7 per cent.
Subnational hunger and undernutrition: The 2030 Agenda emphasizes the importance of using data disaggregated by income, gender, age, and geographic location, among other variables, to ensure that no one is left behind in the development process (UN 2015). While the GHI series provides scores at the national, regional, and global levels, examination of individual GHI indicators at the subnational or state levels reveals disparities within countries, in terms of both absolute values and changes over time. A comprehensive review of subnational differences is not within the scope of this report, nor is it possible given data constraints. Child stunting, child wasting, and child mortality estimates at the subnational level are available irregularly for the countries in this report, and subnational undernourishment estimates are rarely calculated. However, examples of subnational disparities are provided in this section to demonstrate the geographic variation that exists within countries and to serve as a springboard for further research into hunger and undernutrition-and their causes-at the subnational level.
Variations in GHI indicator values can exist within countries at all levels of the GHI Severity Scale, from low to extremely alarming. For countries that have low hunger and undernutrition levels at the national level, examination of data at the subnational level can help identify areas of the country that lag behind. For example, Mexico has a low 2016 GHI score, at 7.2 points, and an overall stunting level of 13.6 per cent, according to the most recent survey data. However, nearly one-third (31.4 per cent) of children in the southern state of Chiapas are stunted (WHO 2013). Child undernutrition in the state has been well documented (Stahl 2014; García-Parra et al. 2016), and Martínez-Rodríguez et al. (2014) find high levels of food insecurity in Chiapas. Also, Gutierrez-Jimenez et al. (2013) show an association between child malnutrition and intestinal parasites in Chiapas. Juarez and Gonzalez (2010) note the lack of proper sanitation in the homes in Chiapas and other states in Mexico, which they suggest may decrease the bioavailability of nutrients-meaning they are not properly absorbed as a result of infection and disease.
Jordan similarly has a low 2016 GHI score at 5.7 points, and its national stunting rate is 7.8 per cent, according to a 2012 survey (WHO 2014). Yet, in the southern state of Ma'an, nearly one-fifth (19.0 per cent) of children under age five are estimated to be stunted (WHO 2014). In fact, Krafft and El-Kogali (2014) show that Jordan has the highest inequality of child stunting rates out of 12 countries in the Middle East and North Africa, while having the lowest national stunting level of these countries.
On the other end of the GHI Severity Scale, examination of the subnational data for the countries categorized as alarming can reveal areas that are in crisis and demonstrate extremely high levels of hunger and undernutrition. Zambia, which has a 2016 GHI score of 39.0 - characterized as alarming - shows substantial differences in terms of the GHI indicators between provinces. According to a 2013-2014 survey, the under-five mortality rate was 7.5 per cent for the nation, but ranged from 6.3 per cent in Copperbelt Province to 11.5 per cent in Eastern Province. To put this in perspective, at the national level, child mortality rates in 2015 are at or exceed 10.0 per cent for only 8 out of 131 countries with data in this report. Zambia's national stunting rate is 40.1 per cent, but is highest in Northern Province, at 48.5 per cent, and lowest in Copperbelt, Lusaka, and Western Provinces, at 36 per cent each. In terms of the wasting rate, the national average is 6.0 per cent, yet Luapula's is more than twice the national average at 13.0 per cent (Zambia, MOH, CSO, and ICF International 2014).
Sierra Leone, also in the GHI's alarming category, has an even wider spread in terms of stunting, whereby its Kono district has the highest stunting level, at 51.6 per cent according to 2013 data, while Bombali district has a stunting level of 28.2 per cent. On the other hand, Bombali has the highest wasting level in the country, at 25.5 per cent, whereas Kono has the second-lowest wasting level, at 4.3 per cent (SSL and ICF International 2014). To address the particular needs of these populations, it is critical to consider the specific circumstances and challenges facing each area.
Cambodia has seen one of the most impressive reductions in hunger between the 2000 and 2016 scores among the countries classified as serious or alarming, dropping by 51 per cent from 44.7 points to 21.7 points. Nationally, the stunting rate fell from 49.2 per cent to 32.4 per cent according to surveys from 2000 and 2014 - a drop of 34 per cent. Yet, some provinces experienced far more impressive reductions in stunting than others. For example, Kandal province saw stunting decline by nearly 50 per cent, dropping from 55.2 per cent to 28.1 per cent in this period. However, in Kampong Speu province, stunting declined by only 18 per cent between 2000 and 2014, going from 49.5 per cent to 40.5 per cent (Cambodia, NIS, DGH, and ICF International 2015; WHO 2012). Similarly, comparing the child mortality estimates from the same surveys, Cambodia's national under-five child mortality estimate fell substantially in the same period, going from 12.4 per cent to 3.5 per cent - a fall of 72 per cent. While several provinces saw their under-five child mortality levels decrease by more than 70 per cent, Kampong Thom's rate fell by only 39 per cent, from 9.9 per cent to 6.0 per cent, which was one of the highest provincial under-five mortality levels in the country (Cambodia, NIS, DGH, and ICF International 2015; Cambodia, NIS, DGH, and ORC Macro 2001). Jimenez-Soto, Durham, and Hodge (2014) examine the 2000, 2005, and 2010 under-five child mortality rates for Cambodia and find persistent, and in some cases increasing, inequality of child mortality in terms of geography.
Conclusion: This year's GHI results demonstrate that the developing world has made great progress in the fight against hunger and undernutrition. At the same time, this report identifies various areas of vulnerability that must be recognized, and data gaps, including at the subnational level, that must be addressed, to ensure that no one is left behind on the path to meeting the goal of Zero Hunger by 2030. In every instance of a region that lags behind the rest of the world, a country that does not keep pace with its neighbours, a country in crisis for which data are inadequate, or a population that is severely disadvantaged within a single country, it is essential that we identify these areas of need and ensure they have the proper support to reach Zero Hunger by 2030.