All our charts on Food Supply
The most widely used and comprehensive data on food supply and consumption is published by the UN Food and Agriculture Organization (FAO). This data is annually available and is updated by the FAO, extending back until 1961. In this chart we see the average daily supply of calories (measured in kilocalories per person per day) by world region, from 1961 to 2013. We have defined the FAO’s dietary energy/protein/fat supply in our Data Quality & Definition section. Note that supply figures do not include consumption-level waste (i.e. that wasted at retail, restaurant and household levels), and therefore represents food available for consumption at the retail level, rather than actual food intake.
Overall, we see that per capita calorie supply has been increasing consistently at a global level over this period. However, these trends vary across the world’s regions. Caloric supply across Europe and Oceania has been approximately constant over this 50-year period, whilst across all other regions, supply has increased. This increase has been most significant across Asia and Africa, whilst in North America, supply has approximately plateaued from 2000 onwards.
The steeper rise across poorer regions of the world means that over the last few decades, global trends in caloric supply have been converging. In terms of food supply, we live in a more equal world today than in the previous century. At the beginning of this entry we have seen the huge gap between the rich and the poor within countries. Now we see that, similarly to what has happened within countries, we are now experiencing a constantly decreasing inequality across the world.
As productivity increased, famines and the exclusion of the poor due to hunger and too little energy became a thing of the past in both France and England. The improving supply of food in both countries is shown in this visualization. In terms of calories, the average food supply in France more than doubled over the last 300 years. Fogel (2004)1, the original source for some of the presented data, brings the numbers in perspective: ‘the energy value of the typical diet in France at the start of the eighteenth century was as low as that of Rwanda in 1965, the most malnourished nation for that year in the tables of the World Bank’.
In this chart we see this FAO data on per capita caloric supply mapped by country from 1961 to 2013.
In 2013, we see that most countries across Europe, Oceania and North America have per capita supplies greater than 3100 kcal/person/day. Countries across South Asia, Sub-Saharan Africa and South America tend to fall within the range of 2300-3000kcal/person/day. However, progress over the last 50 years can be clearly seen by rewinding back to 1961; in this map we see that supplies were more likely to fall within ranges below 2300kcal/person/day across the poorer regions of the world.
It should be noted that there is significant uncertainty with regards to figures published for the USSR prior to its dissolution – figures likely overestimate the level of caloric supply available during this period. The FAO note significant uncertainty in their USSR estimates, in addition to other sources. Detailed discussion on this topic is published by José Luis Ricón (Artir), available here, which we briefly discuss in our section on Data Quality.
You can view the annual trends of any country over time by clicking on it in the map.
Whilst the measure of average caloric supply is a crucial indicator in terms of food security, if we want to relate this to the prevalence of hunger and malnutrition, we must relate it to actual dietary energy needs. In this visualization we have mapped two key variables related to energy requirements through time: the minimum dietary energy requirement (MDER) and average dietary energy requirement (ADER).2
The MDER of an individual is defined as the cut-off threshold caloric intake an individual would have to consume to attain a minimum acceptable weight for height. This metric is used to define the prevalence of undernourishment in a population; an individual whose caloric intake falls below their specific MDER for an extended period is defined as ‘undernourished’. In the map, it is presented as the average MDER for a country’s population which is calculated based on its demographic weighting.
The ADER (sometimes also referred to as the ‘estimated energy requirement’) is defined as the caloric intake required to provide energy balance in a given individual of a healthy weight for their gender, age and activity levels. The ADER is higher than the MDER (reflecting a healthy rather than minimum threshold weight) and is not used in the estimation of undernourishment prevalence.
Why do MDER and ADER values vary across the world? Dietary energy requirements differ by gender and age, and for different levels of physical activity. Accordingly, minimum dietary energy requirements, the amount of energy needed for light activity and minimum acceptable weight for attained-height, vary by country, and from year to year depending on the gender and age structure of the population. So, countries with a very young population may have a lower MDER value on average since the energy requirements of children are typically lower than that of adults. Similarly, countries with a population which generally maintains higher activity levels (for example, with large energy expenditures in laboring or agricultural work) may see an increase in MDER levels.
Whilst there are notable (although narrowing) differences in caloric supply between countries across the world, the variation in supply within countries can, in some cases, be equally significant. To measure the range of caloric intakes in a given population, the FAO define a parameter called the ‘Coefficient of variation of habitual caloric consumption distribution’. In order to assess food security indicators, this has been measured on annual basis for developing countries since 1990. Here we show a global map of this CV value for developing countries.
The CV measures the inequality of caloric intake across a given population. It represents a statistical measure of the data spread around the mean caloric intake. It is measured on a scale from zero to one. where higher CV values represent larger levels of dietary inequality. For example, if a country had a CV value of 0.1, this would represent very equal levels of caloric consumption; most of its population would have a daily caloric intake close to the national mean/average. In contrast, a CV value of 0.5 is determined as high in this context; it represents large dietary inequalities whereby caloric intakes in a given country span a range of values around the national average. An additional metric which is also important to measures of dietary inequality is the ‘Skewness’ (SK) of caloric consumption. SK values define the shape of the caloric distribution around the national average.
In the map we see a range of CV values. Dietary inequality is typically highest in Sub-Saharan Africa (in particular, Zambia, Cote d’Ivoire, and Burkina Faso), Iraq and Haiti. Beyond year-to-year fluctuations, CV values have not drastically changed over time since 1990. South America represents an exception to this trend; it has seen a notable decline in CV (i.e. caloric consumption has become more equal) over time. Brazil’s CV value fell from 0.31 in 1990 to 0.23 in 2014.
The measurement of energy/calories is typically the most standard measure of food supply over time. However, also essential for proper nutrition are our other macronutrients3 Proteins form the building blocks of human tissues, and so protein intake is essential for growth and maintenance.
In this chart we see FAO estimates of per capita protein supply by region from 1961 to 2014. Overall, the global per capita supply of protein has increased by about one-third over this period, rising from 61 grams in 1961 to 81 grams in 2014. Most of this increase has come from growth in Africa, Asia and South America where per capita protein supplies have been rising. Over the past few decades, protein supply in Europe, Oceania and North America have approximately plateaued.
There is, however, a marked difference in protein supplies across these regions. This difference between regions is larger in protein supply than in caloric supply. The per capita supply of protein in North America was 60 percent higher than in Africa in 2014. In terms of calories, this difference was 40 percent. The global distribution of protein supply is, however, narrowing.
In the chart here we see this FAO data on per capita protein supply mapped by country from 1961 to 2013. In 2013, we see that most countries across Europe, Oceania and North America have per capita supplies greater than 100 grams/person/day. Countries across South Asia, Sub-Saharan Africa and South America tend to fall within the range of 50-90 g/person/day. However, increases over the last 50 years can be clearly seen by rewinding back to 1961; in this map we see that most lower-income countries fell within ranges between 30-60 g/person/day. Very few countries in the world had per capita supplies greater than 100 grams per day.
You can view the annual trends of any country over time by clicking on it in the map.
It’s not just the quantity of protein supplies which have changed over this period, but also the composition from different sources of protein. As we will cover in more detail in our entry on Diet Composition over time, our intakes of animal and plant-based foods have been changing. Protein supply can come from both animal and plant sources, however the quality of this protein is not always equal. Typically, animal-based proteins are defined as ‘complete proteins’, meaning they contain all of the amino acid building blocks we need for proper nutrition. Protein found in crops- with some exceptions such as pulses, some nuts and seeds- are often of poorer quality, and lacking in some essential amino acids.4
In this chart we see protein supply differentiated by animal-based and plant-based sources over time. Note that in this chart you can change the country in view. For most countries, we see a general increase in both animal-based and plant-based protein over time. However, if we switch to our “relative” toggle on this chart we see that the share of total protein we get from animal-based sources has been increasing for most countries. We get much more of our protein from meat, dairy and seafood products today than we did 50 years ago.
Fat is one of the three key macronutrients in human diets, alongside carbohydrates and protein. Fats are sources of essential fatty acids, which are important for dietary requirements in a number of ways, including the absorption of vital vitamins, promoting healthy cell function and in provide a buffer against a host of diseases.5
In this chart we see FAO estimates of per capita fat supply by region from 1961 to 2014. Overall, the global per capita supply of fat has increased by over 70 percent since 1961, rising from 48 grams in 1961 to 83 grams in 2014. Unlike calories and protein, fat supply has been increasing across all regions over this period, although this increase has been slowing in North America, Europe and Oceania over the last decade.
Of the three macronutrients, the regional differences in supply are largest for fat. In 2014, the average per capita supply in North America was almost three times as large as in Africa.
In the chart here we see this FAO data on per capita fat supply mapped by country from 1961 to 2013. In 2013, we see that most countries across Europe, Oceania and North America have per capita supplies greater than 140 grams/person/day. Countries across South Asia, Sub-Saharan Africa tend to fall within the range of 20-80 grams, and between 80-100 grams in South America. However, increases over the last 50 years can be clearly seen by rewinding back to 1961; in this map we see that the majority of countries across South America, Sub-Saharan Africa and Asia consumed less than 40 grams per day.
You can view the annual trends of any country over time by clicking on it in the map.
There is a strong correlation between per capita food supplies and prosperity. This is true both between countries (higher-income regions, as we have explored earlier in this entry tend to have higher levels of food supply relative to poorer regions), but also as a country’s prosperity grows over time. As we see in the three sub-sections below, the daily per capita supply of calories, protein and fat all tend to increase with economic growth.
In the sub-sections below we have plotted the relationship between daily per capita food supply (in the form of calories, protein and then fat) versus gross domestic product (GDP) per capita in 2011 international-$. A number of countries selected across a range of income levels and global regions have been highlighted for comparison. For all three measures of food supply, we see a close correlation in per capita supply and prosperity. As we get richer, per capita supply tends to increase.
This growth is particularly true from low to upper-middle income ranges. However, in recent years we have seen an overall plateauing (or in some cases a decline) in per capita supply across a number of high-income countries. For example, if we look at recent trends for the United States, France, the United Kingdom and Japan in the charts below on calories and protein, we see a typical stabilisation or decline in per capita intakes. In some high-income countries this is also true for per capita fat supply, although this plateauing effect is less marked.
Per capita food supplies therefore tend to increase with economic growth, however this rise begins to slow at higher-income levels and in most cases begins to plateau (or even decline) with further growth.
A consequence of increasing per capita food supplies which may seem intuitive is its relationship to the prevalence of undernourishment. Undernourishment, as we cover in detail in our entry Hunger and Undernourishment, is defined as a caloric intake insufficient to meet minimum energy requirements for adequate weight and nutrition.
In this chart we have plotted the prevalence of undernourishment, measured as the percentage of the total population undernourished against the daily supply of calories, in kilocalories per person per day. Overall, we see that countries with lower caloric supplies experience a higher incidence of undernourishment. Over time, we see that as per capita caloric supplies increase, the prevalence of undernourishment decreases. This is shown by a general migration to the bottom-right of the chart.
Note that the FAO does not record levels of undernourishment prevalence below 5 percent; this explains the illusion of a plateauing of prevalence at 5 percent. It notes that: “Data showing as 5 may signify a prevalence of undernourishment below 5%.”
The following definitions are quoted from The State of Food Insecurity in the World6:
- Dietary energy requirement (DER) – The amount of dietary energy required by an individual to maintain body functions, health and normal activity.
- Minimum dietary energy requirement (MDER) – In a specified age/sex category, the minimum amount of dietary energy per person that is considered adequate to meet the energy needs at a minimum acceptable BMI of an individual engaged in low physical activity. If referring to an entire population, the minimum energy requirement is the weighted average of the minimum energy requirements of the different age/sex groups. It is expressed as kilocalories per person per day.
- Maximum dietary energy requirement – In a specified age and sex group, the amount of dietary energy per person that is considered adequate to meet the energy needs for heavy activity and good health. In an entire population, the maximum energy requirement is the weighted average of the maximum energy requirements of the different age and sex groups in the population. This is expressed in kilocalories per person per day.7
- Dietary energy/protein/fat supply – Food available for human consumption, expressed in kilocalories per person per day (kcal/person/day). At country level, it is calculated as the food remaining for human use after deduction of all non-food utilizations (i.e. food = production + imports + stock withdrawals − exports − industrial use − animal feed – seed – wastage − additions to stock). Wastage includes losses of usable products occurring along distribution chains from farm gate (or port of import) up to the retail level. However, such values do not include consumption-level waste (i.e. retail, restaurant and household waste) and therefore overestimates the average amount of food actually consumed.
- Dietary energy supply adequacy – Dietary energy supply as a percentage of the average dietary energy requirement.
- Dietary energy intake – The energy content of food consumed.
As we noted in our section on caloric supply across the world, there is significant uncertainty in figures for the USSR prior to its dissolution. This was first brought to our attention by José Luis Ricón (Artir). He explores this topic in detail here, which we briefly summarize below.
In comparing USSR trends with other countries, a surprising fact emerges: FAO data suggests that the USSR had higher caloric supply than many of the richest countries at the time. In the chart here, for example, we see that the figures suggest food supply was higher in the USSR than the USA until the mid-1980s. If, we compare living standards – take average income for example – this is counterintuitive to what we would expect. During this period the average gross domestic product (GDP) per capita in the USA was around double that of the USSR. It seems unlikely food availability was higher in the USSR than the US.
The FAO acknowledges that their USSR estimates are highly uncertain – here they discuss in detail the sources of this data, doubts of their accuracy and how this affects their estimates methodologically. Why is it suggested that these figures are an overestimate? As Artir discusses in his blog, alongside comparisons with other sources, it is likely that both the quantity and certainly the quality of food consumption in the USSR was below that of the USA. Although in caloric terms it’s expected food supply was sufficient for the majority of the population, it was slightly below that of the USA. And overall quality of diet (in terms of micronutrient availability and dietary diversity) was lower.
How much lower or poorer in quality is hard to say – the data is not available to say with certainty.
Food and Agriculture Organization (FAO) Food Supply database
- Data: Covers food commodities that have been converted back into primary equivalents (Quantity, Dietary Energy, Proteins, Fats, Totals and per Capita). Also includes ‘Per Capita Food Supply Variability’
- Geographical coverage: Global – by country and world region
- Time span: Since 1961
- Available at: Online at FAOSTAT here
- Detailed food consumption data is available in the FAO Food Balance Sheet data. This data is described in FAO (2001) – ‘Food Balance Sheets – A Handbook’ (online here).
- The FAO data is also available through Gapminder where the data can be plotted against a second variable.
The Complex Emergency Database (CE-DAT)
- Data: Mortality and malnutrition data – over 2,000 surveys and 20,000 health indicators
- Geographical coverage: 51 countries
- Time span: 1998 to present
- Available at: Online here
The History Database of the Global Environment (HYDE)
- Data: Total consumption of cereals (wheat and rye) in Europe, in kg/cap and kcal/cap, and Food consumption in the USSR.
- Geographical coverage: Europe and the former USSR
- Time span: 1909-1984
- Available at: Online here
Inequality of Food Consumption
- Data: Coefficient of variation (and skewness) of habitual caloric consumption distribution
- Geographical coverage: Developing countries only.
- Time span: Since 1990.
- Available at: Online here.