1 Introduction

There is growing global consensus of the need to transform food systems in order to achieve critical global goals at the intersection of human and planetary well-being. The Sustainable Development Goals (SDGs) stress that, to meet future needs, we need to use land more sustainably, minimise negative impacts on the environment, and seek opportunities to restore lands that have lost nutrients and/or biodiversity. Simultaneously, it is crucial to provide all people with access to a more nutritious diet, and hence future food systems must provide a diverse range of affordable foods to enable all people to have access to diets of high nutritional quality.

The livestock sector is an important part of these challenges, since, on one hand, it is a major user of land, but, on the other hand, it provides micronutrient-dense food with high-quality protein. Here, we provide a synthesis of the current understanding of the dynamics of the livestock sector in terms of use of natural resources, trade between countries and the synergies and trade-offs caused by the changing nature of the demand and supply of ASF (including milk, meat, eggs, and fish in this study). We discuss the kinds of policies, governance processes and institutions that might minimise negative interactions and maximise positive synergies. We conclude with a brief exposition of the possible implications for the international agricultural research agenda, along with eight priority actions that need to be deployed simultaneously and in combination to ensure that livestock contribute to sustainable food systems, leaving no one behind.

2 Background and Trends

Analyses of trends of the livestock sector suggest that, as incomes increase and societies urbanise, per capita consumption of livestock products increases (Delgado et al. 1999). This, together with increases in population, means that the total demand for livestock products would grow substantially. This phenomenon, while mostly true, hides substantial heterogeneity in terms of the types of livestock products that are likely to increase in demand and the locations of consumption growth. Below, we provide clarity on the dynamics of ASF demand and supply.

2.1 Trends in Animal Source Food Demand: 1990–2015

Averaged globally, over the last 25 years, per capita food demand of all ASF increased by more than 40 kg/person/year (FAOSTAT 2018). However, this number hides substantial variation across regions and by commodity within ASFs, with several different trends operating in opposing directions (Fig. 1). For example, while there was a nearly 35% increase in per capita meat demand (+11.27 kg/person/year) and total per capita meat demand increased for all regions between 1990 and 2015, this increase was driven by large increases in demand for poultry and pork, which saw increases of 106 and 26%, respectively.

Fig. 1
A table represents how demand for animal source foods has changed between 1990 and 2015 as per kilogram or person or year with nine columns of different foods such as fish seafood, milk excluding butter, eggs, meat, bovine meat, mutton and goat meat, pig meat, and poultry meat, with respect to twenty rows of various nations.

Change in animal source food demand in the period 1990–2015 (kg/person/year). (Source: Based on authors’ calculations from FAOSTAT (2018). All regional definitions follow FAOSTAT definitions. Regions are inclusive of selected countries (i.e., Eastern Asia includes China), which are reported individually to highlight key trends)

Global demand for ruminant meat (beef and mutton), however, has followed a different trajectory, with per capita demand having remained near 1990 levels (changing by less than 1 kg/person/year on average globally). Within the beef trend, we still see substantial variation regionally, with most regions exhibiting much bigger declines in beef demand than the global number would suggest.

There is much less diversity of trajectories in the trends for poultry. Per capita poultry demand has increased, with different magnitudes, in all regions. The smallest increase was in Eastern Africa and the United States of America, 27% and 32%, respectively, in per capita demand of poultry meat. All other regions experienced double the per capita demand of poultry meat. Regional pork demand trends are more variable, but resemble poultry more so than beef.

Figure 1 shows the changes in animal source food demand in the period 1990–2015 (kg/person/year) in various regions.

2.2 The Role of Trade in Meeting Demand for Animal Source Foods

The last few decades have seen substantial increases in international trade in ASF, with important regional differences. The value of exports globally has nearly tripled, from around 59 in 1990 to almost 174 billion US$ by 2010, although total trade value represents less than 20% of global production (FAO 2019). Meat, in value terms, has contributed nearly two-thirds of the value of exports of livestock products globally.

Most trade in ASFs is within the same region of origin, with most imports coming from nearby countries, for example, Europe exports to Europe, as shown in Fig. 2. There are, however, a number of dominant trading countries that trade between continents (Fig. 2; for example, intraregional bovine meat exports are dominated by the Southern Cone of South America (most of the green outside of the Latin American region row in Fig. 2), particularly Brazil, Australia (in the East Asian and Pacific region, which is blue), and the United States of America (in the North American region, which is red)). Small ruminant export is dominated by Australia and New Zealand (in the East Asian and Pacific region, which is blue), which are the primary sources of imports for most countries. Europe and, to a lesser extent, North America are the primary exporting regions supplying the bulk of traded intraregional pork. Intraregional trade in poultry is dominated by Brazil (in Latin America, which is green) and the United States of America (in the North American region, which is red).

Fig. 2
A table has 6 columns of headers importing regions, bovine meat, mutton and goat meat, pig meat, poultry meat, and import from 8 different regions.

Composition of 2010 regional imports of meat commodities by source of imports. The source of imports follow the colours given in the final column (i.e., imports from Europe are coloured orange, those from North America are red, etc.), so, for example, 91% of imports of bovine meat in Europe comes from other countries in Europe, whereas 62% of imports of bovine meat in the former Soviet Union comes from countries in Latin America (FAO 2019)

Trade in ASF, in volume terms, is small compared to trade of feed. Trade in meat and processed meat products accounted for less than one tenth of the volume of trade in feed grains (Galloway et al. 2007). These dynamics are likely to intensify as more feed will be necessary to respond to growing demand for pork and poultry in regions currently importing feeds. This comes with substantial consequences for land use and environmental impacts, as, depending on the land used to produce the feed, it could lead to substantial embedded environmental impacts in overall ASF production. A clear example: if imports of soybeans increase in Asia, this could fuel deforestation in Brazil, a primary soybean provider. In other regions, other environmental dimensions would take precedence over emissions, with the potential for substantial losses of biodiversity and disruption of water cycles in places (see Searchinger et al. 2015, for example for Sub-Saharan Africa).

2.3 The Response of Production to Meet the Increase in Demand: The Monogastric “Explosion,” Intensification, and Expansion Dynamics

Since the 1970s, there has been a ‘monogastric explosion,’ with rates of growth in animal numbers often exceeding 4% per year, and meat and egg production, in cases, reaching over 6–7% per year, globally. Improvements in crop yields, improved feeding rations with high-quality feedstuffs, higher production efficiency, favourable prices and the involvement of private industry in driving these dynamics all played a significant role, initially in Europe, North America, and Oceania, and later in Latin America and parts of Asia (FAO 2006).

Since 1990, global production of ASF (kg) has increased by more than 60%, an increase of almost 2% per year (FAOSTAT 2018). Figure 3 shows the regional variations in these patterns. The largest production increases were observed in Africa and Asia. Higher-income regions, on the other hand, grew at a slower rate, with ASF production in Europe actually declining by about 15% from 1990 levels.

Fig. 3
Six bar graphs represent the production of animal products with respect to various global regions, consisting of milk excluding butter, eggs, bovine meat, mutton and goat meat, pig meat, and poultry meat. The production of poultry meat has a maximum value in global regions.

Production trends of animal products (kg) from 1990 to 2015. (Source: Based on authors’ calculations from FAOSTAT 2018)

Across ASF commodities, the fastest growth in production was for poultry meat, which has nearly tripled globally since 1990 (Fig. 3). All regions, on average, saw increased production, with the global median increase in production across all countries being 125% above 1990 levels (~3.3%/year growth).

Eggs, pork, and dairy production grew at a slower pace, with production increasing by 103%, 72%, and 56%, respectively. Beef and lamb production globally grew by about 1/4 and 1/3 the rate of poultry, respectively, since 1990. Lamb production in low-and middle-income regions grew at a much faster rate than the global average, with small ruminant production increasing at rates similar to pork in Africa and Asia. However, in developed countries in North America, Europe, and Oceania, there were declines in production.

Substantial increases in production efficiency, often associated with intensification, have also taken place. Intensification occurred at different rates in different parts of the world and, in some cases, led to reductions in animal numbers. For example, the United States of America produces 60% more milk with 80% fewer cows now than it did in the 1940s (Capper et al. 2009) through a substantial change in genetics, feeding and housing systems. Substantial intensification, as well as expansion, of the livestock sector has occurred primarily in Latin America and Asia. This is in stark contrast with Sub-Saharan Africa, where productivity has remained stagnant for decades, with all of the growth in production being due to increases in animal numbers. These general observations hide substantial heterogeneity, which we disentangle below.

2.4 Different Livestock Products and Production Systems, Different Dynamics

The production increases in the past few decades follow different trajectories for ruminants than for pork and poultry in smallholder or industrial operations (Fig. 4). Between 2000 and 2011, global milk and meat production increased by 28% and 11%, respectively (Fig. 4). Mixed crop-livestock systems contributed to the majority of bovine milk and meat production.

Fig. 4
Four bar graphs of grazing systems and mixed crop livestock systems represent the changes in dairy bovine and meat bovine percentages and the total number of animals by climate dominance and income group. Bovine numbers in low income have the highest value in all four graphs.

Average changes in dairy bovine milk and meat bovine productivities (kg/TLU/year) and animal numbers in grazing systems (a) and mixed crop-livestock systems (b) by climate and income group. Period: 2000–2011. Data calculated based on productivity and animal number estimates by country, livestock system and climate type from (Herrero et al. 2013b). The climate category Arid includes semi-arid systems such as northern Australia. Grazing and mixed crop-livestock systems as defined by (Robinson et al. 2011), income groups as defined by (World Bank 2016). (Figure adapted from Godde et al. 2018)

At the global level, these increases in total production were mainly driven by the increases in animal numbers (dairy: +19%, meat: +10%), followed by the increases in animal productivities (kg of livestock products/TLU/year, milk: +9%, meat: +1%). In arid and humid regions, or in low-income countries, total production increases were mainly driven by the increases in animal numbers, rather than the increases in productivity. This reflects that the feeding systems have remained static, being reliant on animals grazing and harvesting energy from available land, instead of greater utilisation of new forage crops or concentrate feeds. Similarly, improvements in animal health services in these production systems have been limited by patchy disease control, in particular, over remote areas.

In contrast, in temperate regions and high-income countries, total production increases were mainly driven by the increases in productivity, rather than the increases in animal numbers. On average, high-income countries showed a decrease in total animal numbers (−4%) while maintaining modest productivity increases (under 1% per year).

Increases in dairy productivity (28%) only outstripped growth in animal numbers (9%) as the source of growth in dairy production between 2000 and 2011 in the highlands of low-and middle-income countries. This evidence of intensification is unsurprising, considering that the majority of Research and Development and extension efforts have been directed towards these smallholders, mostly mixed dairy systems (Waithaka et al. 2006; Herrero et al. 2010, 2014).

2.5 The Role of Smallholders in the Production of ASF

Livestock production supports about 650 million low-income small-scale producers in lower-and middle-income countries (FAO 2009). Livestock are responsible for 17–47% of the value of agricultural production in the regions of lower-and middle-income countries (Herrero et al. 2013a) and contribute income to 68% of lower-and middle-income country households (FAO 2009), while also playing important cultural roles (Thornton 2010; Herrero et al. 2013a). While men are often most represented in livestock production and fishing, women tend to be highly active in the processing and sale of animal products (Herrero et al. 2013a). At the same time, ASF-related livelihoods do not necessarily entail high-quality jobs. For example, ASF producers and fishing communities in lower-and middle-income countries sometimes do not earn enough to eat their own production (Thow et al. 2017; Annan et al. 2018; Ravuvu et al. 2018). Women in livestock value chains in particular may lack appropriate recognition and remuneration (Agarwal 2018), and denial of women’s access to shared ASF resources, such as fisheries, creates power imbalances that expose women to abuse (Fiorella et al. 2019). A move towards healthier, more plant-rich diets could create more jobs than animal agriculture-based employment, with potential improvements in gender equality and occupational safety (Saget et al. 2020).

Bovine Milk and Meat

Globally, farms smaller than 20 ha produce 45% of bovine milk and close to 37% of bovine meat (Herrero et al. 2017) (Fig. 5). However, important regional differences exist. Large farms (>50 ha) dominate bovine milk (>75%) and meat (>80%) production in North America, South America, and Australia and New Zealand, which are regions with high levels of exports of these products.

Fig. 5
Two horizontal bar graphs represent the production of bovine milk percentage and bovine meat percentage by farm size and global regions. The farm size has a minimum value of two and a maximum value of under 200, respectively.

The production of bovine milk and meat by farm size and region. (Source: Data from Herrero et al. 2017)

Conversely, farms smaller than 20 ha produce the majority (>75%) of bovine milk and meat in China, East Asia Pacific, South Asia, Southeast Asia, Sub-Saharan Africa, and West Asia and North Africa. Very small farms (<2 ha) are of particular importance in China, where they still produce more than 60% of bovine milk and meat. These very small farms are also of importance in East Asia Pacific, South Asia, Southeast Asia, and Sub-Saharan Africa, where they contribute more than 25% of bovine milk and meat production.

The role of smallholders in the future is uncertain. For dairy, a sustainably intensified smallholder sector could be the engine of production growth, as there are still large yield gaps in these systems. Furthermore, with demand both growing and primarily satisfied by local markets (formal and informal), smallholders should benefit from improved cash flow derived from growth in dairy. For intensification to occur, the growth of markets, improved access to inputs and services, and increased adoption of key technological packages need to happen at a faster pace than previously anticipated (McDermott et al. 2010; Godde et al. 2018).

For beef, the situation is different. In the absence of a clear increase in demand per capita, and with small farm output largely dependent on increased numbers of animals, it is likely that operation size will be more of a constraint. Nevertheless, smaller scale production resulting from culled animals in diversified farming systems may continue to be economically viable, even if it is unlikely to be the main source of income or livelihoods.

Pigs and poultry

The contribution of smallholder systems to monogastric production, based on data from Herrero et al. (2013b), shows the importance of smallholder monogastric systems as a source of pork, poultry and eggs in several regions, notably, South and Southeast Asia and Sub-Saharan Africa (Fig. 6).

Fig. 6
A bar graph represents the proportions of pork, poultry, and eggs in the food systems of 10 different regions. It illustrates that South Asia has the highest proportion of poultry.

The proportion of pork, poultry and eggs from smallholder systems in different global regions. (Herrero et al. 2013b)

Gilbert et al. (2015) found a negative relationship between the proportion of extensively raised chickens and pigs and the GDP per capita of different countries. Although there are large variations between countries, this suggests that, as economies grow, the smallholder monogastric sector, while still important in some countries, will tend to reduce in importance as income grows and conditions become more favourable for private industry to industrialise the sector. The reduction in transaction costs and vertical integration will drive this transition, as it has in other regions.

3 Implications of the Historical Supply and Demand Dynamics of ASF for Land Use and Other Environmental Metrics

Livestock account for the majority of greenhouse gas emissions from food systems, through methane from enteric fermentation and manure management, carbon dioxide from land use change, and nitrous oxide from manure management (Herrero et al. 2016; Tubiello et al. 2021). However, livestock now use 62% less land and emit 46% fewer greenhouse gas emissions to produce one kilocalorie compared with 1961. Nevertheless, improved livestock productivity has required an increase of 188% in the use of nitrogen fertilisers derived from fossil fuels to increase feed production (Davis et al. 2015). Despite productivity improvements, due to increased demand, the aggregate environmental impacts of livestock have continued to grow, which will require substantial further reductions in the sector’s environmental footprint.

Animal production practices, depending on type and location, can have beneficial or detrimental effects on biodiversity (Herrero et al. 2009; Barange et al. 2018). In particular, livestock-induced land use conversion is a major environmental and human rights concern in some areas (De Sy et al. 2015). Many intact ecosystems, notably, carbon-dense and biodiversity-rich tropical forest biomes, have been converted to pasture and feed crops for animals (FAO and UNEP 2020). These ecosystems are essential to climate change mitigation (Lennox et al. 2018). Intact ecosystems currently occupy half of the ice-free surface of the Earth (Dinerstein et al. 2017), and this degree of intactness has been proposed as a global limit (Newbold et al. 2016; Dinerstein et al. 2017; Leclère et al. 2018; Willett et al. 2019a, b), implying that an urgent halt to land use conversion is needed. In extensive rangeland practices in grassland and savanna biomes, where large grazers (e.g., bison) have been lost, ruminant livestock can be an important means of biodiversity conservation and climate mitigation (Olff and Ritchie 1998; Griscom et al. 2017).

Resource use varies widely by type of ASF and production practice. Beef production tends to be the greatest user of land and energy, followed by pork, poultry, eggs, and milk production (de Vries and de Boer 2010).

Resource use also varies by production system and setting. In many cases, livestock can be reared on lands of low opportunity cost, without competing with croplands or other land uses (van Zanten et al. 2018). Keeping livestock in grazing systems may have some environmental benefits, such as conservation of grassland biodiversity, although such relationships are complex and context-specific (FAO 2009). Animal production systems are often essential to circular production systems (Poux and Aubert 2018). However, the intensive production of any animal, including pigs and poultry, has substantial environmental impacts, especially for surrounding communities and waterways, that must be considered (Wing and Wolf 2000; Burkholder et al. 2007; Godfray et al. 2018).

3.1 Animal Source Food Consumption Trends: The Three Key Storylines

We review the 2020 projections made by Delgado and others towards the end of the 1990s, contrasted against what is happening currently in the livestock sector. We also summarise the storylines that emerge from these trends. Globally, their projections of total meat and milk production saw a difference of only −12% and −5% from what current trends in FAOSTAT suggest. By commodity, the projections were particularly accurate for pork, with larger deviations for beef and poultry. We observe a similar story with the per capita demand projections. Overall, the projections are good, with a difference of only 4 and 10 kg/person/year for meat and milk, respectively. However, we can see that, similar to the beef and poultry projections, there are offsetting deviations that are masked when we only look at the global number (Table 1). Here, the key deviations are in projections for China and India (Table 2).

Table 1 Comparing global animal source food production (million metric tonnes) in Delgado et al. (1999) to FAOSTAT (2018)
Table 2 Comparing per capita consumption of animal source food (kg/person/year) in Delgado et al. (1999) to FAOSTAT (2018)

Reviewing these projections highlights the fact that the evolution of the global livestock sector over the past couple of decades can be summarised in a few storylines:

  1. (a)

    First, demand for poultry has been the main global driver of increased meat consumption, with per capita consumption having nearly doubled since 1990. This represents a mix of changes in demand and supply.

  2. (b)

    Second, per capita dairy consumption in high-income regions has stayed constant since 1990, with any growth in total consumption being driven by changes in population. Low-and middle-income regions have seen substantial increases in dairy consumption, with this being driven by increases both in population and in per capita consumption of dairy products, with the largest increase observed in China.

  3. (c)

    Finally, increases in global beef demand is a story of two countries, China and Brazil, which account for nearly 93% of the 11 million metric tonne increase in global beef demand, even as, globally, per capita beef consumption has been declining or become stagnant in most countries. The key role of China and Brazil in the global beef sector was already identified by Delgado (2003) in an update of their 1999 projections.

3.2 Animal Source Foods and Human Nutrition and Health: The Need for Moderation, Not Avoidance

In general, healthy plant-rich diets, including flexitarian, or vegetarian options, have lower climate and land impact than those high in ASF; their water and nutrient impacts depend on the practices used (Hallström et al. 2015; Aleksandrowicz et al. 2016; Frehner et al. 2021). Reduction in ASF, notably red meat consumption has been shown to reduce environmental impacts (e.g., on climate, land, and biodiversity), with some studies suggesting that global climate and biodiversity targets are only achievable through reduced consumption (Tilman and Clark 2014; Leclère et al. 2018; Springmann et al. 2018; Clark et al. 2020). For example, transition to healthy plant-rich diets that include some meat could reduce food-related emissions by nearly half, setting them on track to meet the 1.5 °C climate target (Clark et al. 2020). In contrast, a global transition to increased consumption of ASF, notably red meat, is not feasible within recommended environmental limits (Springmann et al. 2018).

Diets that include few or no ASFs, including vegetarian and vegan diets, have been shown to reduce the risk of non-communicable diseases (Tilman and Clark 2014; Springmann et al. 2016). Diets with diverse plant-sourced foods can meet protein requirements (Young and Pellett 1994), and vegetarian diets can meet adult micronutrient needs (Walker et al. 2005). However, plant-based foods do not necessarily equal healthy foods: many highly processed foods are fully plant-based (e.g., highly processed snack foods and sugar-sweetened beverages), yet have been associated with poor health outcomes (Hu 2013; Marlatt et al. 2016; Mozaffarian 2016).

Controversy exists regarding dietary recommendations for some ASF, and this has had a polarising effect on many scientific and food sector discussions. Confounding this debate are statements regarding global calls to reduce ASF consumption (Willett et al. 2019a, b), masking regions where increased intake would have positive impacts on health; creating confusion in the health impacts of a diversity of ASF; causing a lack of clarity (and sometimes unrealistic assumptions) on which foods would be replacing ASF in diets; and, finally, leading to the under-consumption of health-promoting foods (Afshin et al. 2019).

While ASF consumption and its subtypes are highly variable geographically, the under-consumption of whole grains, fruits, nuts and seeds, vegetables, and seafood, together with excess sodium, remains the largest risk of disease and mortality attributed to diets, according to the Global Burden of Disease Dietary Risk Factors study (Afshin et al. 2019). For many in high-income settings, increasing the consumption of protective foods while remaining within caloric recommendations may require reduced consumption of some ASF. For others, particularly resource-constrained populations in low-and middle-income countries, increasing consumption of certain ASF (alongside consumption of these other protective foods) could have health benefits. However, there are limitations within the underlying data, uncertainty regarding these estimates and significant heterogeneity in consumption among subpopulations (Beal et al. 2019).

3.3 Animal Source Foods and Undernutrition

ASFs are considered complete sources of protein that provide all nine essential amino acids in adequate quantities and are the only dietary source of vitamin B12. In addition, ASFs are nutrient dense and have higher bioavailability of key nutrients such as iron, vitamin A, and zinc compared to plant source foods, although nutrient content may vary depending on the type of ASF. Consumption of these foods may be particularly essential for young children, adolescent girls, and pregnant or lactating women, as these individuals have increased nutrient requirements due to biological processes (Neumann et al. 2002; Murphy and Allen 2003; Semba et al. 2016; Beal et al. 2017). With regards to undernutrition, a number of studies have assessed the role of ASFs in linear growth for children under the age of five and micronutrient deficiencies in both women and children. Recent systematic reviews have identified limited evidence regarding the association between consumption of ASF and linear growth during early childhood (Eaton et al. 2019; Pimpin et al. 2019; Shapiro et al. 2019). These reviews concluded that substantial heterogeneity in definitions of ASFs might have led to inconsistent results. On the other hand, a cross-sectional analysis of Demographic Health Surveys found a strong association between ASF consumption and reduced incidence of stunting, and consumption of a diversity of ASF had an additive effect on that relationship (Headey et al. 2018). Another study found a strong correlation between ASF intake and reductions in stunting in Nepal and Uganda, with dairy consumption having the strongest correlation (Zaharia et al. 2021). In addition, in a longitudinal sample of children in rural Nepal, increased ASF consumption was associated with greater child development scores (Miller et al. 2020). Of the ASF, small pelagic fish, molluscs, large pelagic fish, salmonids and carp tend to have higher nutritional density than most terrestrial ASF, with ruminant meats following (Golden et al. 2021). A diversity of foods, including of ASF, within healthy ranges remains a standard recommendation, particularly for nutritionally vulnerable populations.

3.4 Animal Source Foods and the Risk of Non-communicable Diseases

The relationship between ASF and chronic diseases is highly dependent on the type of ASF and what other foods are substituted for ASF in the diet (e.g., red meat vs. other lean protein or red meat vs. ultra-processed plant-based foods). Cohort studies provide modest evidence that increased consumption of low-fat dairy and seafood may be protective against cardiovascular disease (Bernstein et al. 2010; Soedamah-Muthu et al. 2011; Schwingshackl et al. 2017a, b). On the other hand, the association between unprocessed and processed red meat consumption and diet-related chronic diseases is still debated among scientists. In 2015, the International Agency for Research on Cancer classified processed meat as a group 1 carcinogen (other examples: tobacco smoking and outdoor air pollution) and unprocessed red meat as a probable carcinogen (IARC 2015). There is strong evidence to suggest that consumption of processed meat (cured, salted, preserved) is associated with increased risk of cancer on average (Bouvard et al. 2015), although the precise mechanisms and differences among subtypes require more study. The relationship between unprocessed (fresh) red meat and health is more controversial and needs further research. Some epidemiological cohort studies have found positive associations between unprocessed red meat consumption and, respectively, overall mortality (Schwingshackl et al. 2017a, b; Zheng et al. 2019), type-2 diabetes (Pan et al. 2011; Schwingshackl et al. 2017a, b), cardiovascular disease (Qian et al. 2020), and cancer (Chan et al. 2011; Bouvard et al. 2015), while other studies have not found any relationship between unprocessed red meat and adverse health outcomes (Johnston et al. 2019; Iqbal et al. 2021). While more research is needed, most studies that do suggest higher adverse health risk with red meat consumption find an association at doses exceeding 1–2 servings/week (Mozaffarian 2016), which is consistent with many national dietary recommendations (Gonzalez Fischer and Garnett 2016; Herforth et al. 2019; USDA and USHHS 2020).

In summary, populations consuming high amounts of red meat, particularly in processed forms, would benefit from reduced consumption to improve health and sustainability. This mostly applies to consumers in higher-income countries, but also to a growing number in lower-and middle-income countries, where the burden of diet-related non-communicable diseases is growing rapidly. For those vulnerable to undernutrition (whether in lower-and middle-income countries or higher-income countries), the nutrient contribution of minimally processed ASF may be beneficial in reducing risk of micronutrient deficiency and promoting growth (Murphy and Allen 2003).

4 Essential Actions for Ensuring Livestock’s Contribution to Sustainable Food Systems

This section examines some alternative or additional actions that would need to take place for livestock to contribute to sustainable food systems, while addressing critical aspects of social equity, poverty and other social goals. As discussed, this will require different actions depending on the context, including:

  • Consumption of ASF at a level appropriate to meet nutritional needs.

  • A reduction in consumption of red and processed meat for populations with high risks of diet-related non-communicable diseases or in the context of an unbalanced diet.

  • The enabling of increased consumption by nutritionally vulnerable populations that need higher levels of nutrients, including pregnant women, the elderly, children and undernourished populations, particularly those in lower-and middle-income countries.

Several studies have quantified the potential environmental gains of changing dietary patterns. This area of work started from the need to quantify the greenhouse gas mitigation potentials of changing diets (Stehfest et al. 2009), and has been expanded considerably to include health impacts and several additional environmental metrics (Tilman and Clark 2014; Leclère et al. 2018; Springmann et al. 2018; Willett et al. 2019a, b). As an example, Fig. 7 summarises the technical mitigation potential of changing diets.

Fig. 7
A horizontal bar graph represents the change in diets versus greenhouse gas mitigation potentials. The vegan diet, which includes no animal source food has a peak value of eight, and the Mediterranean diet, which includes moderate meat but is rich in vegetables, has at least a value of three.

The technical greenhouse gas mitigation potential of changing diets according to a range of scenarios examined in the literature. (Mbow et al. 2019)

The features of these studies show that:

  1. 1.

    The upper bound of the technical mitigation potential of demand-side options is about 7.8 Gt CO2-eq per year (no consumption of animal products scenario) (Stehfest et al. 2009).

  2. 2.

    Many dietary scenario variants have been tested. Key variants include target kilocalorie levels (i.e., 2500 kcal per capita per day), notions of healthy diets, swaps between animal products (red vs. white meat) and/or vegetables, and stylised diets (Mediterranean, flexitarian, etc.). All fit roughly between the current emissions and the Stehfest et al. (2009) upper boundary.

  3. 3.

    The main impact of reducing the consumption of ASFs is to reduce the land footprint of livestock. This land-sparing effect, coupled with alternative uses of the land (i.e., negative emissions technologies), leads to a large mitigation potential. Many of the other environmental impacts are also associated with the land-sparing effect (i.e., biodiversity, Leclère et al. 2018).

  4. 4.

    The largest technical potential comes from reductions in ruminant meat consumption (the most inefficient sub-sector), as most scenarios try to trigger land sparing (reduction of carbon dioxide emissions) as the key mechanism for reducing emissions.

  5. 5.

    Reductions in livestock product consumption, especially red meats, could have both environmental and health benefits (Tilman and Clark 2014; Willett et al. 2019a, b).

  6. 6.

    Fully vegan diets could meet calorie and protein requirements, but can also be deficient in key nutrients (vitamin B12, folate, Zinc), a concern for vulnerable groups, in particular, those without access to dietary supplements. Therefore, diets with some level of animal products may be necessary.

  7. 7.

    The economic mitigation potential of changing diets is not known. This is a crucial research area, together with mechanisms for eliciting behavioural changes.

  8. 8.

    Most scenarios so far have taken kilocalories as the currency for changing diets; few have dealt with protein or micronutrients, a factor that, from the livestock and healthy diet perspectives, seems like a necessary step.

  9. 9.

    Very few key examples of legislation and policy-induced shifts in consumption exist. There are some examples that have been shown to promote increases in the consumption of fruits and vegetables (Garnett et al. 2015).

  10. 10.

    The social and economic costs of reduced demand for ASFs are unknown. Notably, there is little information on the impacts on farmers’ income, employment, alternative labour markets, reductions in agricultural GDP, etc.

  11. 11.

    Methodological advances are needed to elicit simultaneously the environmental, health and socio-economic impacts of reduced consumption.

Attached to livestock production is an enormous amount of wealth generation, employment, value chains and famers’ livelihoods. Impacts on these are seldomly studied, yet they are crucial to creating convincing policy cases for a contraction of livestock product demand. Global studies that have begun to include some of these critical feedbacks are only now starting to emerge (Mason-D’Croz et al. 2020).

From a nutritional perspective, livestock’s contribution to healthy diets is not so much about their kilocalories as their micronutrients and protein. It is essential to include these in future research. Diets in these scenarios are also too ‘globalised,’ and more realistic, and culturally sensitive, regional variants will need to be examined. The differentiated impacts of ASF consumption and production across population cohorts will require that future analysis begin to better recognise the heterogeneity of populations (rural/urban, under-or over-nourished, gender, age, or by age groups), if they are to provide necessary information to improve the targeting of future food policies.

Mitigating greenhouse gases from livestock systems is more feasible in some contexts than in others, and this largely depends on the livelihood objectives of livestock farmers (Herrero et al. 2016). Nonetheless, many practices that improve productivity or the production system as a whole can lead to direct and indirect greenhouse gas mitigation co-benefits. These should be pursued.

The supply side options for mitigating greenhouse gases in the livestock sector have been the subject of recent reviews (Smith et al. 2007, 2014; Hristov et al. 2013; Herrero et al. 2016; Roe et al. 2019). These options look to:

  • Reduce the enteric methane of ruminants.

  • Reduce nitrous oxide through manure management of both ruminants and monogastrics.

  • Implement best animal husbandry and management practices (all), which would have an effect on major greenhouse gases (carbon dioxide, methane and nitrous oxide).

  • Directly sequester carbon from pastures (ruminants)

  • Generally improve land use practices that also help in enhancing soil carbon sequestration.

Excluding land use practices, Herrero et al. (2016) found that these options have a technical mitigation potential of 2.4 GtCO2eq/year. However, they also found that the economic feasibility of these practices is low (10–15% of the technical potential, or less than 0.4 GtCO2eq/year). The largest mitigation opportunities for the livestock sector occur when livestock are considered holistically as part of the agriculture, forestry and land use sectors (Havlík et al. 2014).

5 Concluding Remarks and Recommendations in the Context of the Food Systems Summit

Our study has demonstrated that the dynamism of the livestock sector provides a range of avenues for change, some more relevant to smallholders than others, some more amenable to public funding than others, and some more likely to alleviate negative environmental impacts than others. Picking the most effective and desirable solutions will be essential for stakeholders associated with the livestock sector to achieve the desired impacts on sustainable food systems. The balance between social and environmental goals will need to be carefully evaluated. The avenues for growth, the trade-offs and the potential actions can be summarised below.

Smallholder dairy: The evidence suggests that demand for milk is growing fast, and that, at least in highland or high potential areas, productivity per animal is increasing due to the adoption of better practices, like feeds, animal health management and genetics. These systems can be competitive, but issues surrounding land fragmentation and feed availability need closer attention. Testing and implementing transformational feed technologies or engaging in developing systems that could increase in circularity, such as through increased biomass recycling, sound like important next steps to ensure high-quality feed at low environmental costs in these systems. This needs to go beyond previous work on crop residues (e.g., Blummel and colleagues) and may need transdisciplinary partnerships with other sectors to develop these new biomass streams and to adjust breeding and feeding strategies. This, in turn, would also lead to reduced pressures on land and to the exploration of other greenhouse gas mitigation avenues, beyond those explored to date (improved feeds, manure management). Eventually, this could contribute to the national mitigation action plans of specific countries.

The smallholder pork and poultry sector: our synthesis has shown that, while there are countries where smallholder pork and poultry makes an important contribution to the supply of these products, in the coming decades, much of the growth in production is likely to come from industrial production, as integrated supply chains emerge and private sector engagement increases. This suggests that investing in these smallholder systems is, at best, a medium-term strategy that could provide livelihood benefits as these producers diversify or identify new exit strategies. Identifying transition options for these producers in the future seems necessary.

From an international public good perspective, the future of feed in fuelling the large demand for pork and poultry is a critical researchable issue, if the feed is to be sourced sustainably. Biomass value chains, old and new, need to be evaluated, developed and promoted to ensure that competition for food with humans is minimised. Here, again, circular feed sources, regulations for including a minimum amount of recycled feed, and new feed sources (superfeeds from industrial production or others) need to be developed, along with professional arguments in favour of local industries taking on these enterprises in a well-planned manner.

For monogastrics, there are a lot of researchable issues, including on antimicrobial resistance, with priority areas being:

  1. 1.

    Monitoring inputs: what inputs are used in the system in terms of feed, antimicrobials and other aspects that affect the health of the animals and have implications for the health of producers, consumers and those working in the food chain.

  2. 2.

    Surveillance: establishing systems that generate information on current and emerging diseases, antimicrobial use and antimicrobial resistance.

  3. 3.

    Assessment: the economic burden of livestock health and wealth (see https://animalhealthmetrics.org) as a basis to identify interventions that impact positively on the economic outcomes of livestock production, as well as minimising impacts on the environment and public health.

A central element of a livestock agenda in relation to environmental trade-offs is related to the identification of entry points for engaging in the beef sector. On one hand, the existing data show that most of the growth in red meat production has been obtained through increases in animal numbers, while intensification has been influential in only a few countries. Consumption per capita is stagnant, or decreasing, in most countries, and most of the demand is driven by population growth. At the same time, reducing red meat consumption could lead to substantial greenhouse gas mitigation, as well as reductions in pressure on land and biodiversity. Producing red meat only on lands of low opportunity costs, or as a by-product of the dairy industry, would have the lowest environmental impacts.

Identifying the best levels of consumption in relation to other dietary elements for different population groups should be a high priority for the Food Systems Summit, as well as identifying ways to decouple red meat production from land, or create niche products for very specific sets of consumers through labelling systems and certification.

The livestock sector will change, voluntarily, or as a result of forces external to it. If sustainability concerns are of paramount importance, a critical research area is to develop economic incentive systems (price premiums) and regulations to pay for reduced emissions, watershed protection, biodiversity protection and others; and to internalise these in true cost or true-pricing schemes, supported by adequate regulatory and fiscal measures.