Subsistence farming and how families grow food to feed themselves
Subsistence farming is a livelihood in which a family grows crops or keeps animals chiefly to eat, buys few inputs and sells a minor part of what it harvests, in a definition cited by Morton (2007). How many households farm this way is not known, because no global count exists.
Definitions from labour statisticians and farm economists come first, then three farming systems, the regions where they persist and the effect of home-grown food on household diets. Agricultural censuses compiled by United Nations agencies, satellite mapping, national household surveys and reviews in peer-reviewed journals supply the evidence. A closing section separates measured findings from estimates.
What subsistence farming is
A widely cited definition comes from Barnett and colleagues and is quoted by Morton (2007): the household eats the main output, buys few if any inputs and markets a small proportion of the harvest. Published definitions are scarce. Morton (2007) calls their number surprisingly small, given how widely both subsistence agriculture and smallholder agriculture are used as terms.
FAO (2014) draws the line by purpose, not by size. Its typology places subsistence or near-subsistence smallholders in a separate group whose families produce essentially for their own consumption and have little or no chance of a surplus for the market.
Labour statisticians use a stricter wording. Under the 2013 ILO resolution, subsistence foodstuff producers grow, catch or gather food that contributes to the livelihood of the household, and people who do so as recreation are excluded (ILO, 2013). The working test is simple. The work is done without workers hired for pay or profit (ILO, 2013).
Subsistence and commercial farming side by side
The ILO (2013) standard separates the two groups by where the output is meant to go. Goods intended mainly for own consumption stay in the subsistence group even when a part or a surplus is sold or bartered. Farmers who produce mainly for sale count as employed, even if the household eats part of the output (ILO, 2013).
Real farms rarely sit at either pole. Morton (2007) places smallholder and subsistence farmers on a continuum running from subsistence production to concentration on crops for the market. Household surveys from Africa show the same pattern. Market participation is widespread there, yet commercialisation remains limited (Christiaensen, 2017).
Carletto and colleagues measured the crop commercialisation index in Malawi, Uganda and Tanzania, as summarised by Christiaensen (2017). On average, farmers sold around 20–25% of their crop output by value, a bit less in Malawi and slightly more in the other two countries. About three quarters of the output by value was therefore not sold, an inference from that average and not a measured share eaten at home.
Types of subsistence farming
Shifting cultivation
Shifting cultivation is defined by the balance between cropping and rest. In the forestry journal Unasylva, Lanly (1985) recorded the operational definition: relatively short periods of continuing cultivation followed by relatively long periods of fallow. When the fallow lasts more than twice as long as the cropping period, the system is sometimes called long-fallow agriculture (Lanly, 1985). His worked example used 5 years of cultivation and 15 years of fallow (Lanly, 1985).
Heinimann and colleagues (2017) mapped the system from satellite data and expert surveys. They put shifting cultivation landscapes at roughly 280 million hectares worldwide, counting cultivated fields together with fallows, and they call the figure an approximation. Africa holds the largest share of that area, ahead of the Americas and Asia (Heinimann et al., 2017).
Resting land between crops is also the starting idea of [how crop rotation works](/agriculture/how-crop-rotation-works/), where the sequence of crops changes instead of the field. The system persists where population density is low and other options for farming or income are limited (Heinimann et al., 2017).
Nomadic pastoralism
Pastoral households live mainly from herds. FAO (2001) describes exclusive pastoralists as producers who grow no crops and obtain food by selling or exchanging animals and their products. Many herders follow a fixed seasonal circuit. Transhumance, in the FAO (2001) wording, is the regular movement of herds between fixed points to use pastures as they become available through the year.
Agropastoralists sit between herding and farming. FAO (2001) calls them settled pastoralists who cultivate enough land to feed their families from their own crops. Nearly all herders still buy staple food. Morton (2007) notes that pastoralists almost all depend on selling livestock and livestock products to buy staples and other necessities.
Nobody has counted them precisely. A joint evaluation published by IFAD (2016) gives estimates of 50 million to 200 million pastoralists worldwide.
Intensive subsistence farming
Rice offers the clearest example of small plots farmed season after season. FAO (2016) describes rice in Asia and in Africa as mainly a small farmer crop, with almost all of it grown on holdings of 0.5 to 3 hectares.
Plots of that size can be productive. In the same guide, FAO (2016) reports that one hectare of paddy farmed for rice and fish together can yield up to 9 tonnes of rice and 750 kilograms of fish a year, a maximum and not an average. Fertiliser choice changes how much such small paddies return, a question examined in [fertiliser types and rice yield under SRI](/agriculture/fertilizer-types-rice-yield-soil-properties-sri-agriculture/).
Where subsistence farming is practised today
Small farms in numbers
No comprehensive global count of subsistence farming households is available. Morton (2007) found few informed estimates of their world or regional population and noted that FAO does not publish data split into these categories. Farm censuses come closest. Lowder and colleagues (2019) estimate that the world has more than 608 million farms.
Most of those farms are tiny. Farms under 1 hectare make up 70% of all farms but operate only 7% of agricultural land (Lowder et al., 2019). China alone holds 34% of the 608 million farms and India 24% (Lowder et al., 2019).
Size alone does not identify a subsistence farm. An earlier working paper by the same lead author (Lowder et al., 2014) points out that a small farm may supply a large share of the owner's income or food. Its owners may be subsistence farmers in poverty, or they may earn enough to support the family.
Regions and land
Asia dominates the farm count. Of the 570 million farms in the FAO (2014) estimate, almost 75% were in Asia. Only 9% were in sub-Saharan Africa, 7% in Europe and Central Asia and 4% in Latin America and the Caribbean (FAO, 2014).
Land is shared differently in poorer countries. In low and lower-middle income countries, about 80% of farms are under 2 hectares and they operate about 30 to 40% of the land (Lowder et al., 2019).
Europe still has a subsistence tail. Of the 8.8 million agricultural holdings in the EU in 2023, one third (2.9 million) had a standard output below €2,000 a year and produced only 1% of EU agricultural output (Eurostat, 2026). Eurostat places these very small farms at the semi-subsistence end of the scale, a classification by economic size, not a measured share of food eaten at home.
What growing its own food gives a household
Own production and diet diversity
The hope behind diverse home plots is a more varied family diet. Jones (2017) reviewed 23 studies on the question. Seventeen used data from sub-Saharan Africa, and nearly all were cross-sectional snapshots with a single longitudinal analysis (Jones, 2017). In 19 of the 21 studies that measured diet diversity, more agricultural biodiversity went with more diverse diets, though in most the association was small (Jones, 2017).
Ecuador gives a direct example of the home-grown route. Households whose farms held more species ate more food items from their own production (Jones, 2017).
A meta-analysis by Sibhatu and Qaim (2018) gives a measure of size. It pooled 45 studies from 26 countries and found that less than 20% reported consistently positive and significant links between production diversity and diets or nutrition. Its mean effect of 0.062 implies that a farm would need 16 extra crop or livestock species to add one food group to the diet, and around 9 in Africa (Sibhatu and Qaim, 2018). Child growth shows a similar picture: Jones (2017) puts the association with height-for-age at about 0.05 standard deviations.
A buffer when markets fail
Home production matters most where buying food is hard. In one Mexican study cited by Jones (2017), greater crop diversity went with less dependence on purchased food, an advantage where food prices were higher than wages. Jones (2017) adds that edible species may bring more nutritional benefit than cash crops where food markets are thin or prices are high relative to wages.
The same review urges caution about cause. Eating more home-grown food may be a coping response of poorer households to food insecurity, not a benefit of farm diversity (Jones, 2017). Only six studies measured market access, and in all six the link with diet diversity held after adjustment, but causal inference from these studies remains limited (Jones, 2017). Sibhatu and Qaim (2018) found little evidence that diversifying production is a highly effective way to improve smallholder diets in most situations.
How much of the world’s food small farms produce
The best-known figure concerns family farms, a broader group. FAO (2014) estimates that family farms occupy around 70–80% of farmland and produce more than 80% of the world's food in value terms. The same report warns that its farm estimates indicate orders of magnitude rather than exact numbers (FAO, 2014).
Family farms include large commercial holdings. Most small farms are family farms, but not all family farms are small, Lowder and colleagues (2019) stress. The largest 1% of farms operate more than 70% of the world's farmland (Lowder et al., 2019).
Two estimates compared
Farms under 2 hectares produce roughly 36% of the world's food, by the estimate of Lowder et al. (2019). Those farms make up 84% of all farms worldwide and operate around 12% of agricultural land. These estimates describe different measures of farm size, land use and food production, and neither directly counts the food consumed by subsistence farming households (Lowder et al., 2019).
Study | Data and coverage | What was measured | Main finding |
FAO (2014) | Agricultural censuses from different years and countries | Food produced by family farms, by value | More than 80% of world food |
Lowder et al. (2014) | Census data behind the 2014 FAO estimate | Farm numbers and land by farm size | 84% of farms under 2 ha on about 12% of farmland |
Lowder et al. (2019) | Country food values weighted by land share | Food produced by farms under 2 ha | Roughly 36% of world food |
Eurostat (2026) | EU farm structure statistics for 2023 | Standard output of holdings | 2.9 million holdings below €2,000 produced 1% of output |
Christiaensen (2017) | Household surveys in Malawi, Uganda and Tanzania | Crop sales as a share of production value | Around 20–25% of crop output sold |
Note: these figures describe family farms, small farms and very small holdings, which overlap with subsistence farms without matching them. None of them counts households that eat most of what they grow.
Risks of farming mainly for the household
Weather and climate shocks
Households that eat their own harvest absorb a bad season directly. Morton (2007) describes these systems as complex, diverse and risk-prone, with small farms often held under informal tenure in marginal environments. Warming adds pressure on staple yields. In the tropics, even moderate temperature rises of 1–2 °C are likely to reduce yields of rice, maize and wheat, according to the evidence reviewed by Morton (2007).
Model projections reported in the same review are uneven. Modelling cited by Morton (2007) projects that maize yields in smallholder rain-fed systems in Africa and Latin America will fall by about 10% by 2055, and that average hides large local variation. Morton (2007) lists small farm size, low technology and low capital as sources of vulnerability. He adds that family labour, existing diversification away from farming and indigenous knowledge give resilience that should not be underestimated.
Reliance on a few staples concentrates that exposure. Rice, maize and wheat are the three major cereals worldwide and among smallholder and subsistence farmers (Morton, 2007). Pests add a further risk where one tree crop supplies much of a household's food, as the case of [breadfruit mealybug control on Pacific atolls](/agriculture/breadfruit-mealybug-biological-control-pacific-atolls-agriculture/) shows.
Poverty and market access
Farming for the household rarely lifts it out of poverty on its own. FAO (2014) found that most subsistence farms depend heavily on non-farm income and are unlikely to escape poverty through agriculture alone. Hunger is concentrated in the same places. More than 70% of the world's food-insecure people live in rural areas of developing countries, and many are farm labourers or subsistence producers (FAO, 2014).
Tools and water are scarce. In the African household surveys reviewed by Christiaensen (2017), 32% of cultivating households owned and 12% rented some farm equipment, less than 1% owned a tractor and less than 5% used any water control. FAO (2014) expects farmers in the subsistence category to have little capacity for a marketable surplus. They are unlikely to be drawn into effective agricultural innovation systems.
What is proven about subsistence farming and what is not
Subsistence farming is well defined in principle and poorly measured in practice. The definitions agree on purpose, but the counts available describe farms by size or by family ownership.
Small farms and family farms can be counted, but no source counts the households that live mainly on what they grow.
Several points are established. Farms under 2 hectares make up 84% of farms on about 12% of farmland (Lowder et al., 2014). Farms under 1 hectare make up 70% of farms on 7% of land (Lowder et al., 2019). The ILO (2013) gives an operational test: food production for the household without hired workers.
Other findings depend on definitions. Pastoralist numbers depend on the definition used and are not known with any precision (IFAD, 2016). Food-share estimates rest on land share as a proxy for output (Lowder et al., 2019). Even the farm total is likely an undercount, according to Lowder and colleagues (2014).
Some questions remain open. Nobody knows how many people depend on shifting cultivation today (Heinimann et al., 2017). The same team estimates that the system will tend towards disappearance in all regions by 2090, a tentative projection (Heinimann et al., 2017).
Morton (2007) expects smallholder and subsistence households to decline in number as they move into other livelihoods. National statistics add a blind spot of their own: macro numbers do not account well for production for own consumption, which in developing countries makes up a substantial share of agricultural output (Christiaensen, 2017).
Frequently asked questions
Do subsistence farmers still exist?
Yes, although no source counts them directly. In the EU, 2.9 million holdings with a standard output below €2,000 a year sat at the semi-subsistence end of the scale in 2023 (Eurostat, 2026). In the tropics, shifting cultivation landscapes still cover roughly 280 million hectares including fallows (Heinimann et al., 2017).
What is semi-subsistence farming?
It describes households that eat most of their harvest but sell part of it. Statistical standards allow for this: food intended mainly for own consumption stays in the subsistence group even when a surplus is sold or bartered (ILO, 2013). In Malawi, Uganda and Tanzania, farmers sold around 20–25% of their crop output by value (Christiaensen, 2017).
Is subsistence farming the same as smallholder farming?
No. Smallholder refers to farm size, while subsistence refers to where the output goes. A small farm may provide a large share of income, and its owners may be poor subsistence farmers or may earn an adequate living (Lowder et al., 2014). Most small farms are family farms, but not all family farms are small (Lowder et al., 2019).
What do subsistence farmers grow?
Cereals dominate. Rice, maize and wheat are the three major cereals worldwide and among smallholder and subsistence farmers (Morton, 2007). In Asia and Africa, almost all rice comes from holdings of 0.5 to 3 hectares (FAO, 2016). Herding households keep animals instead of fields and exchange animals or their products for food (FAO, 2001).
Why is subsistence farming common in Africa and Asia?
Small farms are concentrated there. Almost 75% of the world's farms were in Asia in the FAO (2014) estimate. In low and lower-middle income countries, farms under 2 hectares operate about 30 to 40% of the land (Lowder et al., 2019). Shifting cultivation lasts where population density is low and other livelihoods are limited (Heinimann et al., 2017).
Sources
- Morton JF. The impact of climate change on smallholder and subsistence agriculture. Proceedings of the National Academy of Sciences of the United States of America. 2007;104(50):19680–19685. DOI 10.1073/pnas.0701855104.
- FAO. The State of Food and Agriculture 2014: Innovation in family farming. Rome: Food and Agriculture Organization of the United Nations; 2014. ISBN 978-92-5-108536-3. https://fao.org/3/i4040e/i4040e.pdf
- Lowder SK, Skoet J, Singh S. What do we really know about the number and distribution of farms and family farms in the world? ESA Working Paper No. 14–02. Rome: FAO; 2014. https://fao.org/3/i3729e/i3729e.pdf
- Lowder SK, Sánchez MV, Bertini R. Farms, family farms, farmland distribution and farm labour: What do we know today? FAO Agricultural Development Economics Working Paper 19–08. Rome: FAO; 2019. ISBN 978-92-5-131970-3. https://fao.org/3/ca7036en/ca7036en.pdf
- Heinimann A, Mertz O, Frolking S, et al. A global view of shifting cultivation: Recent, current, and future extent. PLOS ONE. 2017;12(9):e0184479. DOI 10.1371/journal.pone.0184479.
- Lanly JP. Defining and measuring shifting cultivation. Unasylva. 1985;37(147):17–21. https://fao.org/4/r1340e/r1340e04.htm
- FAO. Pastoralism in the new millennium. FAO Animal Production and Health Paper 150. Rome: FAO; 2001. ISBN 92-5-104673-5. https://fao.org/3/Y2647E/y2647e02.htm
- IFAD. FAO’s and IFAD’s Engagement in Pastoral Development: Joint Evaluation Synthesis. Document EC 2016/92/W.P.6. Rome: IFAD; 2016. https://webapps.ifad.org/members/ec/92/docs/EC-2016-92-W-P-6.pdf
- FAO. Save and Grow in practice: maize, rice, wheat. A guide to sustainable cereal production. Rome: FAO; 2016. ISBN 978-92-5-108519-6.
- Christiaensen L. Agriculture in Africa - Telling myths from facts: A synthesis. Food Policy. 2017;67:1–11. DOI 10.1016/j.foodpol.2017.02.002.
- Eurostat. Farms and farmland in the European Union - statistics. Statistics Explained. European Commission; 30 April 2026. https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Farms_and_farmland_in_the_European_Union_-_statistics
- International Labour Organization. Resolution concerning statistics of work, employment and labour underutilization. Adopted by the 19th International Conference of Labour Statisticians. Geneva: ILO; 2013.
- Sibhatu KT, Qaim M. Review: Meta-analysis of the association between production diversity, diets, and nutrition in smallholder farm households. Food Policy. 2018;77:1–18. DOI 10.1016/j.foodpol.2018.04.013.
- Jones AD. Critical review of the emerging research evidence on agricultural biodiversity, diet diversity, and nutritional status in low- and middle-income countries. Nutrition Reviews. 2017;75(10):769–782. DOI 10.1093/nutrit/nux040.
Related
How crop rotation works and why it improves soil fertility
How fertiliser types affect rice yield and soil under SRI
Why breadfruit mealybug control failed to last on Pacific atolls
Why Punjab Farmers Burn Stubble and What Fire Counts Show
How Much Can New Woodland Cut UK Emissions? What the Models Show
No-till farming and what long-term field trials show about yields
What GMO means and what research shows about genetically modified food