Methodology
Calculating equivalent number of animals spared through reduced consumption of animal-based meals.
Since eating patterns around the world vary substantially, the impact of altering dietary behaviours or meal consumption is also likely to vary between countries, regions, and continents. The per-capita consumption of meat might be quite similar in two different regions, but one region might consume more pork while the other consumes more poultry – thus a meat-reduction intervention might spare more animals in the latter country. Furthermore, there are other differences that have an impact on eating patterns: for example, the yield per animal (meat, milk, eggs) might be quite different in different regions, again impacting the number of animals needed to produce a certain quantity of meat, milk, or eggs. Last but not least, the consumption of animal products per person might be low in some regions, while total production might be quite high due to a large population.
When calculating the number of animals spared by replacing meat-centred meals with plant-based ones, we assume that every meal we replace was not plant-based beforehand. We further assume that every meal replaced with a more plant-friendly option contains about 100 g of meat. Furthermore, for the ProVeg Grants Program, our current calculations use a global average.
In order to take all of the above into account, we use data on production and consumption provided by the FAO as they provide a comprehensive overview of all countries, regions, and continents. Note that we have chosen 2016 as the baseline year for our calculations.
We used the following data for our calculations:
- ‘Livestock Primary’: this data set contains data on production in tons as well as the number of producing or slaughtered land animals per region per year. (1)
- ‘Food Balance Sheets’: this data set contains information on production, imports, exports, and other utilisations, as well as the food supply per person per year for specific products. (2)
- ‘Detailed Trade Matrix’: this data set contains data on the trade of different products between different countries. (3)
Calculation steps
- Calculate the yield per animal from domestic production
- Calculate the edible yield per animal
- Calculate the number of animals used for each species in 100 g of meat
- Calculate the average number of animals per meal
In our country- and regional-level calculations, there is an additional step between steps 1 and 2 that corrects the yield per animal for imports, with the help of trade data.
- Calculate the yield per animal from domestic production
To calculate the yield per animal, we simply divide the amount of meat produced in a certain region by the number of animals slaughtered. Note that production refers to the dressed carcass weight, excluding offal and slaughter fats but including bones.
(production (tons) 1000) / animals slaughtered=yield per animal (kg)
Example: pig meat, globally
- Production: 118.8 million tonnes
- Animals slaughtered: 1.48 billion Animals
(118,787,713 t 1000)/1,480,033,104 animals=80.26 kg/animal
- Calculate the edible yield per animal
As data on production is expressed in terms of dressed carcass weight, it includes bones and other parts that are not generally consumed by humans. In order to correct for this, we use a couple of factors to calculate the edible meat yield:
- 70% for meat from cattle
- 75% for pig, sheep, and goat meat
- 80% for meat from poultry
In the case of pig meat, this would be:
80.26 kg per animal x 0.75 = 60.2 kg per animal
- Calculate the number of animals in 100g of meat per species
In the next step, we calculate the number of animals in 100 g of meat (and since we assume 100 g of meat per meal, the number of animals per meal). This is done by dividing 100 g by the edible meat yield.
Example, for pig meat: 0.1kg/60.2 kg per animal = 0.00166 animals in 100g of meat
This is done for all the animal species for which we provide calculations. These numbers can be used to calculate the number of animals spared when replacing meals that contain specific kinds of meat, e.g. poultry meat or meat from cows.
- Calculate the average number of animals per meal
Because we generally don’t know exactly what kinds of meals we influence, we thus need to generalise. As each country has its own specific dietary patterns, the composition of an ‘average’ meal can vary quite widely and contain different animals. After we calculate the number of animals in 100 g of meat per species, we calculate a weighted average, weighing it against the per-capita food supply of those items in each region.
Table: Food supply, with different kinds of meat and numbers of animals in 100g of meat (global average)
| item | Food supply (kg/capita/year) | Animals per 100g/Animals per meal |
| Meat from cattle | 8.88 | 0.00066 |
| Meat from sheep and goats | 1.97 | 0.0092 |
| Pig meat | 15.37 | 0.00166 |
| Poultry meat | 15.31 | 0.0730 |
| Other meat* | 0.75 | 0.044 |
| Weighted average | 0.0284 |
*’Other meat’ contains the meat of rabbits, horses, mules, camels, other rodents, other camelids, and others.
Result for the ProVeg Grants Program (as of August 2021)
Given that the ProVeg Grants Program has influenced 45.8 million meals thus far, this equates to a total of about 1.3 million land animals spared. (Data as of August 2021.)
Limitations
This is a not comprehensive list of limitations.
Limitations of the FAO data
Data from the different databases provided by the FAO can, from time to time, contain slightly different data on the same area of analysis. This can vary to such a degree that the production values from the production data do not always fit the production data from the Food Balance Sheets, and that import and export data from the Detailed Trade Matrix do not always directly match the trade data from the Food Balance Sheets. Furthermore, data from the FAO might sometimes differ from official country statistics.
Data that is not included
The calculations currently reflect only meat consumption. Dairy products and eggs have not yet been considered. Furthermore, the model currently excludes fish and invertebrates. If these animals were included, the number of animals spared could increase tremendously. Furthermore, the data currently being used only reflects animals used directly for human consumption – it does not include, for example, animals that have died before slaughter or the number of male chicks that have been culled during egg production.
Edible meat yield
For now, only one figure per animal is used and applied globally. Of course, this might not be completely accurate in every case. As with consumption and yields, the utilisation of the carcass and leftovers is also likely to vary between regions. For our calculations and the sake of simplicity, we currently assume the same value for every region. This might be adjusted in the future.
Meat cuts and what is mostly eaten from the animal
The number of animals consumed per meal could be further influenced by other factors that we haven’t covered in this more general calculation. For example, even though we calculated the edible meat yield per animal, there might also be specific cuts from animals that are used for meals, such as breasts or wings from chickens, ham from pigs, or ribs from cattle.
By way of example: using our current calculation, we assume that 1,000 meals (not specified) replaced in the US equates to about 1,000 x 0.0254 = 25.4 animals. However, if we know that all of those meals contained poultry meat, the number of animals required would be about 1,000 x 0.0534 = 53.4 animals.
References
1.Food and Agriculture Organization of the United Nations (2021): Livestock primary. FAOSTAT Database. Rome, Italy. Available at: http://www.fao.org/faostat/en/#data/QL [04.04.2021]
2.Food and Agriculture Organization of the United Nations (2021): Food Balances (2014- ). FAOSTAT Database. Rome, Italy. Available at: http://www.fao.org/faostat/en/#data/FBS [03.08.2021]
3.Food and Agriculture Organization of the United Nations (2021): Detailed trade matrix. FAOSTAT Database. Rome, Italy. Available at: http://www.fao.org/faostat/en/#data/TM [03.08.2021]
Since eating patterns around the world vary substantially, the impact of altering dietary behaviours or meal consumption is also likely to vary between countries, regions, and continents. The per-capita consumption of meat might be quite similar in two different regions, but one region might consume more pork while the other consumes more poultry – thus a meat-reduction intervention might spare more animals in the latter country. Furthermore, there are other differences that have an impact on eating patterns: for example, the yield per animal (meat, milk, eggs) might be quite different in different regions, again impacting the number of animals needed to produce a certain quantity of meat, milk, or eggs. Last but not least, the consumption of animal products per person might be low in some regions, while total production might be quite high due to a large population.
When calculating the number of animals spared by replacing meat-centred meals with plant-based ones, we assume that every meal we replace was not plant-based beforehand. We further assume that every meal replaced with a more plant-friendly option contains about 100 g of meat. Furthermore, for the ProVeg Grants Program, our current calculations use a global average.
In order to take all of the above into account, we use data on production and consumption provided by the FAO as they provide a comprehensive overview of all countries, regions, and continents. Note that we have chosen 2016 as the baseline year for our calculations.
We used the following data for our calculations:
- ‘Livestock Primary’: this data set contains data on production in tons as well as the number of producing or slaughtered land animals per region per year. (1)
- ‘Food Balance Sheets’: this data set contains information on production, imports, exports, and other utilisations, as well as the food supply per person per year for specific products. (2)
- ‘Detailed Trade Matrix’: this data set contains data on the trade of different products between different countries. (3)
Calculation steps
- Calculate the yield per animal from domestic production
- Calculate the edible yield per animal
- Calculate the number of animals used for each species in 100 g of meat
- Calculate the average number of animals per meal
In our country- and regional-level calculations, there is an additional step between steps 1 and 2 that corrects the yield per animal for imports, with the help of trade data.
- Calculate the yield per animal from domestic production
To calculate the yield per animal, we simply divide the amount of meat produced in a certain region by the number of animals slaughtered. Note that production refers to the dressed carcass weight, excluding offal and slaughter fats but including bones.
(production (tons) 1000) / animals slaughtered=yield per animal (kg)
Example: pig meat, globally
- Production: 118.8 million tonnes
- Animals slaughtered: 1.48 billion Animals
(118,787,713 t 1000)/1,480,033,104 animals=80.26 kg/animal
- Calculate the edible yield per animal
As data on production is expressed in terms of dressed carcass weight, it includes bones and other parts that are not generally consumed by humans. In order to correct for this, we use a couple of factors to calculate the edible meat yield:
- 70% for meat from cattle
- 75% for pig, sheep, and goat meat
- 80% for meat from poultry
In the case of pig meat, this would be:
80.26 kg per animal x 0.75 = 60.2 kg per animal
- Calculate the number of animals in 100g of meat per species
In the next step, we calculate the number of animals in 100 g of meat (and since we assume 100 g of meat per meal, the number of animals per meal). This is done by dividing 100 g by the edible meat yield.
Example, for pig meat: 0.1kg/60.2 kg per animal = 0.00166 animals in 100g of meat
This is done for all the animal species for which we provide calculations. These numbers can be used to calculate the number of animals spared when replacing meals that contain specific kinds of meat, e.g. poultry meat or meat from cows.
- Calculate the average number of animals per meal
Because we generally don’t know exactly what kinds of meals we influence, we thus need to generalise. As each country has its own specific dietary patterns, the composition of an ‘average’ meal can vary quite widely and contain different animals. After we calculate the number of animals in 100 g of meat per species, we calculate a weighted average, weighing it against the per-capita food supply of those items in each region.
Table: Food supply, with different kinds of meat and numbers of animals in 100g of meat (global average)
| item | Food supply (kg/capita/year) | Animals per 100g/Animals per meal |
| Meat from cattle | 8.88 | 0.00066 |
| Meat from sheep and goats | 1.97 | 0.0092 |
| Pig meat | 15.37 | 0.00166 |
| Poultry meat | 15.31 | 0.0730 |
| Other meat* | 0.75 | 0.044 |
| Weighted average | 0.0284 |
*’Other meat’ contains the meat of rabbits, horses, mules, camels, other rodents, other camelids, and others.
Result for the ProVeg Grants Program (as of August 2021)
Given that the ProVeg Grants Program has influenced 45.8 million meals thus far, this equates to a total of about 1.3 million land animals spared. (Data as of August 2021.)
Limitations
This is a not comprehensive list of limitations.
Limitations of the FAO data
Data from the different databases provided by the FAO can, from time to time, contain slightly different data on the same area of analysis. This can vary to such a degree that the production values from the production data do not always fit the production data from the Food Balance Sheets, and that import and export data from the Detailed Trade Matrix do not always directly match the trade data from the Food Balance Sheets. Furthermore, data from the FAO might sometimes differ from official country statistics.
Data that is not included
The calculations currently reflect only meat consumption. Dairy products and eggs have not yet been considered. Furthermore, the model currently excludes fish and invertebrates. If these animals were included, the number of animals spared could increase tremendously. Furthermore, the data currently being used only reflects animals used directly for human consumption – it does not include, for example, animals that have died before slaughter or the number of male chicks that have been culled during egg production.
Edible meat yield
For now, only one figure per animal is used and applied globally. Of course, this might not be completely accurate in every case. As with consumption and yields, the utilisation of the carcass and leftovers is also likely to vary between regions. For our calculations and the sake of simplicity, we currently assume the same value for every region. This might be adjusted in the future.
Meat cuts and what is mostly eaten from the animal
The number of animals consumed per meal could be further influenced by other factors that we haven’t covered in this more general calculation. For example, even though we calculated the edible meat yield per animal, there might also be specific cuts from animals that are used for meals, such as breasts or wings from chickens, ham from pigs, or ribs from cattle.
By way of example: using our current calculation, we assume that 1,000 meals (not specified) replaced in the US equates to about 1,000 x 0.0254 = 25.4 animals. However, if we know that all of those meals contained poultry meat, the number of animals required would be about 1,000 x 0.0534 = 53.4 animals.
References
1.Food and Agriculture Organization of the United Nations (2021): Livestock primary. FAOSTAT Database. Rome, Italy. Available at: http://www.fao.org/faostat/en/#data/QL [04.04.2021]
2.Food and Agriculture Organization of the United Nations (2021): Food Balances (2014- ). FAOSTAT Database. Rome, Italy. Available at: http://www.fao.org/faostat/en/#data/FBS [03.08.2021]
3.Food and Agriculture Organization of the United Nations (2021): Detailed trade matrix. FAOSTAT Database. Rome, Italy. Available at: http://www.fao.org/faostat/en/#data/TM [03.08.2021]
