A healthy diet costs in global average US$3.68 per day in 2021. This is considerably higher than the average food expenditure in almost all low income countries, where people have to do with a diet dominated by staples and oils, lacking protein and a number of micronutrients. The cost of healthy food is also out of reach for many people living above the World Bank’s extreme poverty line.
A team of researchers have summarized data of the cost of a global ”Healthy food basket” which is closely aligned to a set of national public dietary guidelines. The healthy food basket is composed of six broad categories of food, starchy crops, oils, vegetables, fruits, legumes, nuts and seeds and animal foods. The researchers selected a small number of the cheapest food items in each category to calculate minimum cost of a healthy diet in the countries of the world. Notably, the healthy food basket has a considerably higher share of fruits and vegetables and a lower share of animal source foods (13% compared to 18%) than the average global consumption.
It is well established that a healthy diet can be costly. Fruits, vegetables and animal foods are relatively expensive while starchy foods, oils and legumes are much cheaper when calculated in cost per energy unit. Oils give in average almost 20 times more food energy than vegetables. You eat vegetables for dietary variation, taste and micronutrients rather than to get sated. Meat is mostly eaten for protein and fat as well as taste and status. Still, meat gives much cheaper energy than both fruits and vegetables. The cost is also the reason for why poor people, mostly, eat very little greens, fruits and animals foods and a lot of staple foods like grain and root crops (see for instance in Pradhan et al 2013). Oils and other fats were traditionally in very short supply in most parts of the world, and most of the fat was of animal origin. The extreme expansion of vegetable oils, mostly palm oil and soy oil, has now made fat cheaper than starch as an energy source.
The research is based on market prices and obviously both actual consumption and ”cost” will be different for the considerable part of the global poor that are farmers themselves. But even for them I believe it is clear that the poorest often eat a very starch-based diet, supplemented by purchased vegetable (palm) oil, sugar and salt (obviously there are some poor populations engaged in fisheries, livestock or the collection of wild plants that have a different diet). I worked many years in very poor countries and with poor farmers in East Africa and it was striking how seldom they consumed any quantity of vegetables. Those who did grow them did it mostly for the market, to get very much needed income.
Smallholder Susan Mkandawire in Zambia cooks maize porridge with some salt, palm oil and pumpkin leaves to her family. She sells most of the vegetables she grows and the few chickens she raises to get much needed cash. Photo: Richard Mulonga
The World Bank now defines extreme poverty by an income below $3 in PPP US dollars (Purchasing Power Parities) in low-income countries, $4.20 in lower-middle income countries and $8.30 in upper-middle income countries. The actual research is for the situation 2021, before most of the recent food price hikes. At that time, the World Bank poverty limit was $2.15. Realistically, around $1.35 of that could be used for food, which is more than one third of the cost of a healthy diet. The World Bank estimates that a little less than half of the global population live with a daily income below $8.30 in 2025. Considering that the bank increased the poverty line in upper-middle income countries from $6.85 to $8.30 between 2022 and 2025 it seems like a fair estimate that half of the global population can’t afford a healthy diet. This also calls into question of how the poverty lines are defined. Shouldn’t a person above the poverty line be able to eat healthy food?
The researchers conclude that the so called Eat-Lancet diet is even more costly than the Healthy Diet Basket used in this study. This is due to the fact that the EAT Lancet diet has more categories of food and specify quantities of expensive foods such as nuts and fish. One can really question the relevance of making recommendations such as the Eat-Lancet diet when it is out of reach of most people. The Health Diet Basket would, according to this research cause slightly more greenhouse gas emissions than the Eat Lancet.
Of course, one can’t draw too far reaching conclusions from this kind of research. In the end food is about a lot more than cost and calories and even if national data has been used, food consumption data is not particularly accurate and even less so in countries with high levels of self-provisioning.
What constitutes a healthy diet is also vigorously debated and I don’t want to get into details about it here. My own opinion is that a mixed diet based on local foods will be fine, which means bigger variations than in the efforts to prescribe global diets. Where I live in the Sweden it will mean a diet with more animal foods than the global average and less fruit. In almost no countries of the world, people eat as much vegetables as is recommended in dietary recommendations and despite being a passionate vegetable grower since 1977 I am not convinced about the feasibility of increasing vegetable consumption to satisfy nutritional recommendation (Rundgren 2019).
The research referred to here is not fine grained enough to cover all aspects, as the researchers point out themselves. In many cases the starch component will be refined (or be sugar) and not be whole grain for instance, and fish, various meats and dairy have different nutritional profiles and health reputation.
In a coming article I will look into to what extent countries can feed their population with a healthy diet.
References
Herforth, A.W., Bai, Y., Venkat, A. et al. The Healthy Diet Basket is a valid global standard that highlights lack of access to healthy and sustainable diets. Nat Food 6, 622–631 (2025). https://doi.org/10.1038/s43016-025-01177-0
Our World in Data 2025, Share in poverty relative to different poverty lines, World, https://ourworldindata.org/grapher/share-in-poverty-relative-to-different-poverty-thresholds
Pradhan P, Reusser DE, Kropp JP (2013) Embodied Greenhouse Gas Emissions in Diets. PLoS ONE 8(5): e62228. https://doi.org/10.1371/journal.pone.0062228
Rundgren, G 2019, Five dollars a day is not enough for five a day, Garden Earth https://gardenearth.substack.com/p/five-dollars-is-not-enough-for-five-day
Rundgren, G 2022, Food and agriculture number crunching, part 3, Garden Earth
World Bank 2025, Measuring Poverty. https://www.worldbank.org/en/topic/measuringpoverty
Photo: Richard Mulonga. Author suppliled.
