Posted on February 6, 2012, 5:31 PM
A vegetarian spares the lives of a certain number of animals each time he or she chooses to forgo meat for vegetables, fruits, grains, legumes and nuts. But, exactly how many animals does a vegetarian save each year? Given the scale and complexity of animal agriculture today, this number is impossibly difficult to determine accurately. But, it is possible to estimate a conservative number, say X, to allow a claim that a vegetarian saves at least X number of animals. In this post, I will attempt such an estimate for a vegetarian in the United States.
- Vegetarian Resource Group. How Many Adults Are Vegan in the U.S.?. (link, accessed February 22, 2012)
- US Census Bureau. Population Estimates. (link, accessed February 22, 2012)
- USDA, National Agricultural Statistics Service. Poultry Slaughter 2010 Summary. February 2011. (link, accessed January 1, 2013)
- USDA, National Agricultural Statistics Service. Livestock Slaughter 2010 Summary. April 2011. (link, accessed January 1, 2013)
First, a few preliminaries. To determine the number of animals saved by a vegetarian, we need at least two numbers: the total number of animals killed for food consumed in the US in a given year and the size of the US population during that year. But, estimating the number saved is not merely a matter of dividing the total number killed by the size of the population. Let me illustrate this with a simple example. Suppose there are only three people in the US: one regular meat-eater who eats 100 animals each year, one semi-vegetarian who eats 50 animals each year and a vegetarian who eats 0 animals each year. A reasonable conclusion is that the semi-vegetarian saves 50 animals per year and the vegetarian saves 100 animals per year. But, if we merely divide the number killed by the population size, we will unreasonably conclude that a vegetarian saves only 150/3 = 50 animals per year. Similarly, if we merely divide the number killed by the size of the meat-eating population, we will again unreasonably conclude that a vegetarian saves only 150/2 = 75 animals per year.
So, it is important to consider the semi-vegetarians and the Meatless Mondayers, especially since their numbers are quite large, and give them appropriate credit in our calculations. If we say that a vegetarian saves X animals per year, we should be able to also say that a semi-vegetarian saves X/2 animals per year and that a Meatless Mondayer saves X/7 animals per year. So, we will use the following formula to estimate the number of animals saved.
|Number saved by a vegetarian =||Total number of animals killed|
|Population size ∗ ( 1.0 − V − S/2 − M/7 )|
where V is the fraction of the population that is vegetarian, S is the fraction of the population that is not vegetarian but semi-vegetarian (defined as those who eat vegetarian at more than half their meals) and M is the fraction that is neither vegetarian nor semi-vegetarian but does eat vegetarian at least one day per week (such as a Meatless Mondayer). Using the results of the most recent poll commissioned by the Vegetarian Resource Group, I will use V as 0.05, S as 0.16 and M as 0.04.
In the following, almost all of the data for the number of animals killed is for the year 2010. So, using data from the U.S. Census Bureau, I will use the resident population of the United States on July 1, 2010 (mid-year) as 309.33 million.
The number of land animals killed in the U.S. or imported minus the number exported determines the U.S. supply of meat in the market. The following table lists these numbers. The table does not include cows used for dairy, calves used for veal, hens used for their eggs or the male chicks killed by the egg industry because we are trying to find the number of animals saved by a vegetarian, not a vegan (a whole other topic for another post, another time). Two annual reports produced by the USDA serve as our sources for the number who are slaughtered or condemned: the Poultry Slaughter 2010 Summary report and the Livestock Slaughter 2010 Summary report. The import and export numbers come from the data on U.S. livestock and meat trade provided by the USDA here. Conversion from carcass weights or pounds of meat to numbers of animals is based on data in the Poultry Slaughter and Livestock Slaughter reports mentioned earlier.
|Slaughtered in the U.S. or imported||Exported from U.S.||Estimated deaths before slaughter||Total killed for U.S. supply of meat|
- USDA, Animal and Plant Health Inspection Service. APHIS Info Sheet: Veterinary Services. May 2010. (link, accessed February 22, 2012)
- Agrosoft, Ltd. Understanding the Relationship between Performance and Cost. (link, accessed February 22, 2012)
- G. T. Tabler, I. L. Berry and A. M. Mendenhall (University of Arkansas, Division of Agriculture). Mortality Patterns Associated with Commercial Broiler Production. Avian Advice 6(1), 2004. (link, accessed February 22, 2012)
- USDA, Economic Research Service. The Economic Organization of U.S. Broiler Production. June 2008. (link, accessed February 22, 2012)
- M. Petracci et al. Preslaughter Mortality in Broiler Chickens, Turkeys, and Spent Hens Under Commercial Slaughtering. Poultry Science 85(9), September 2006. (link, accessed January 1, 2013)
- University of California, Avian Sciences Department. California Turkey Production. September 1997. (link, accessed February 22, 2012)
In the table above, the estimated number of deaths before the moment of slaughter are based on the ante-mortem condemnation rates reported by the USDA, estimated mortality rates on the farm and estimated deaths during transportation. The mortality numbers for bovines from causes other than slaughter come from a 2010 USDA report on the mortality rates on U.S. beef operations. Mortality numbers for pigs come from a study involving over 50,000 sows conducted by Agrosoft, Ltd., a consultancy company on pig business management. The number of chickens who die in the broiler industry before reaching the age of slaughter is based on mortality patterns associated with commercial broiler production described in a publication by researchers in the Poultry Science Department at the University of Arkansas. The “harvested” age used for this estimate is based on a 2008 USDA publication on the economic organization of U.S. broiler production. Chickens and turkeys die in large numbers during transport to the slaughter plant and this number comes from the pre-slaughter mortality rates reported here. Finally, the rate at which turkeys die before they reach slaughter comes from a report on California turkey production by the Kearney Agricultural Center at UC Davis.
We kill over 8.1 billion land animals each year for our food. Using the formula mentioned earlier, we find that a vegetarian saves over 30 land animals each year, over 28 of them being chickens.
Note: This post was updated on February 12, 2012, with data on import/export of land animals and deaths of land animals in the industry before reaching the age of slaughter. As a result, the number changed from 34 to 30.
- National Marine Fisheries Service. Fisheries of the United States: 2010. August 2011. (link, accessed February 22, 2012)
- A. Mood. Worse Things Happen at Sea: The Welfare of Wild-Caught Fish. 2010. (link, accessed February 22, 2012)
- National Oceanic and Atmospheric Administration, Fisheries Service. Aquaculture in the United States. February 2012. (link, accessed February 22, 2012)
- National Oceanic and Atmospheric Administration and USDA. The Future of Aquafeeds. December 2011. (link, accessed February 22, 2012)
Aquatic animals we buy for food come to us through at least three different means: commercial landings (caught in the wild, brought ashore and then sold), aquaculture (farmed aquatic animals) and imports. The sum of these minus the exports yields the total that enters the U.S. supply for sale as food. Using data from the Fisheries of the United States 2010 report by the National Marine Fisheries Service (NMFS), Fisheries Statistics Division, the following table shows these numbers for 2010 (aquaculture number is for 2009) in thousands of pounds of live weight (i.e., full weight of the animals as caught). The number under landings represents edible domestic commercial landings (edible implies it does not include fish caught and used for industrial purposes such as fish oils, fertilizers, fish meal and shellfish meal; domestic implies it does not include fish caught in international waters or off U.S. shores; commercial means it does not include fish caught in recreational activities).
(aquaculture number is for 2009)
(in thousands of pounds)
Now, the table above only gives us tonnages but not the actual number of fish killed for our consumption. To deduce the number of fish, we will rely on a report produced by fishcount.org.uk, which painstakingly estimates the mean weights of well over 400 species of fish. Using the tonnages reported by the Food and Agriculture Organization of the United Nations (FAO), the report concludes that a sum total of 970 to 2,700 billion fish caught worldwide weigh a total of about 170.61 billion pounds. Being conservative in our estimate of the number of fish, we will use the lower end of their estimate (970 billion). The average live weight of a fish caught to serve human consumption worldwide, therefore, is about 0.176 pounds (we catch a lot of anchovies!). But, why is the worldwide catch of fish relevant to us? According to the National Oceanic and Atmospheric Administration (NOAA) within the U.S. Department of Commerce, 86% of the seafood we eat is imported and so, the worldwide catch of fish is representative of who we kill for what we eat.
- Food and Agriculture Organization of the United Nations. Fishery and Aquaculture Statistics: 2009. 2011. (link, accessed February 22, 2012)
- Food and Agriculture Organization of the United Nations. Indicative Factors for Converting Product Weight to Live Weight for a Selection of Major Fishery Commodities. (link, accessed February 22, 2012)
- Counting Animals. An Estimate of the Lower Bound on Wild Fish Killed to Serve Human Consumption in the US. Research Notes of Counting Animals, Note 1, February 22, 2012. (link, accessed February 22, 2012)
- J. Bostock et al., Aquaculture: Global Status and Trends. Philosophical Transactions of the Royal Society B 365: 2897–2912, 2010. (link, accessed February 22, 2012)
According to the NOAA, 43% of sea animals we eat is an imported product of aquaculture and an additional 5% is a domestic product of aquaculture. Since the majority of what we import are fish products (as opposed to shellfish products), we can conservatively say that at least 48% of the fish we eat is a product of aquaculture. So, 48% of 8,510,523 thousand pounds, i.e., 4,085,051 thousand pounds of live weight of fish we eat is a product of aquaculture. The fishcount study estimates that the average live weight of a farmed fish is between 0.5 pounds to 11 pounds. To be super-conservative, I will use the high end of their estimate (11 pounds). So, this implies that U.S. residents eat over 371 million aquacultured fish. Using the formula described earlier, a vegetarian saves at least 1.3 aquacultured fish.
But, it is important to not forget the fish we catch, then kill and process into something to feed to the fish we farm. For every pound of a farmed fish product, more than one pound of wild catch is used as fishmeal or fish oil to feed the farmed fish. According to the recently released NOAA-USDA report, The Future of Aquafeeds, the overall fish-in fish-out ratio in modern fish aquaculture today is 1.5:1. So, we multiply 4,085,051 thousand pounds with 1.5 to get 6,127,576 thousand pounds as the live weight of fish we kill to feed the farmed fish we eat. Adding to this the rest which is not a product of aquaculture (52% of 8,510,523 thousand pounds, i.e., 4,425,472 thousand pounds), we get a total of 6,127,576 + 4,425,472 = 10,553,048 thousand pounds of wild fish caught to serve human consumption (either directly for us to eat or through being fed to farmed fish we eat).
According to the Food and Agriculture Organization's 2009 yearbook, 46% of the worldwide food supply is from aquaculture. This is slightly smaller than the 48% of the U.S. food supply that is a product of aquaculture. Therefore, since the average live weight of a fish used to feed other fish is much smaller than that of fish we catch and eat ourselves, it is mathematically sound to divide the U.S. tonnage by the average live weight of a fish caught worldwide in order to obtain a conservative estimate of the total number of wild fish killed to serve human consumption in the U.S. So, 10,553,048 divided by 0.176 pounds gives us 59.96 billion fish. Using the formula discussed at the beginning of this post, we find that a vegetarian saves at least 224 wild fish.
This implies that a vegetarian saves at least 1 aquacultured fish (large fish weighing about 11 pounds) and 224 wild caught fish (weighing an average of 0.176 pounds). So, a vegetarian saves over 225 fish each year.
Doesn't this number of fish seem too high? No, because we are counting the number of fish we kill to serve human consumption and not just the number of fish that pass through our lips. The following example illustrates how easy it is for a person to cause the death of over 225 fish a year.
Suppose I eat a small 6-ounce piece (0.375 pounds) of a farmed salmon fillet (a serving size used in an official Weight Watchers' recipe). According to the U.N. Food and Agriculture Organization's conversion factors, this is equivalent to 0.75 pounds of live weight of a salmon. According to authors at the Institute of Aquaculture at the University of Stirling, the fish-in-fish-out ratio for salmon is as high as 4.9. So, about 3.6 pounds of live weight of wild caught fish would have to be killed and ground up to make the fishmeal used to feed and grow the salmon in my meal. The average weight of these wild-caught fish is conservatively estimated at 0.176 pounds (see discussion above). Now, 3.6 divided by 0.176 is 20.4. So, just eating one serving of a farmed salmon fillet from a Weight Watchers' recipe kills as many as 20 fish. It is, sadly, too easy to end up killing over 225 fish per year!
Note: The fish section of this post was updated on February 22, 2012, because the NOAA updated its web site with new numbers related to U.S. aquaculture. The total number of fish saved by a vegetarian changed from 219 to 225.
The shellfish we buy for our food also comes to us through three means: commercial landings, aquaculture and imports. The sum of these minus the exports yields the total tonnage of shellfish entering the U.S. supply for consumption. Using data from the Fisheries of the United States 2010 report again, the following table shows these numbers for 2010 (aquaculture number is for 2009) in thousands of pounds of live weight (i.e., full weight of the animals as caught including their shells).
(aquaculture number is for 2009)
(in thousands of pounds)
In the table above and in the rest of this post, I convert the weights reported in pounds of frozen or fresh meat (without including the shell), canned meat or breaded meat into live weights using conversion factors mentioned or implied in the Fisheries of the United States 2010 report (if available) or those suggested by the Food and Agriculture Organization of the United Nations here.
- Department of Commerce. Imports and Exports of Fishery Products: Annual Summary, 2010. (link, accessed February 22, 2012)
- J. S. Foer. Eating Animals. Back Bay Books, 2010.
- G. Matheny. Least Harm: A Defense of Vegetarianism from Steven Davis’s Omnivorous Proposal. Journal of Agricultural and Environmental Ethics 16:505-511, 2003. (link, accessed February 22, 2012)
So, we eat over 4.5 billion pounds of shellfish in live weight. To estimate the number of animals in these 4.5 billion pounds, we need a breakdown of this number by species type and then use estimates of the mean weights of each. I use the Fisheries of the United States 2010 report for the breakdown of the landings and aquaculture numbers. I use the Imports and Exports of Fishery Products Annual Summary 2010 report produced by the Department of Commerce for the breakdown of the import and export numbers. All of this tabulation leads to the following table, which says that 55% of the live weight of shellfish we eat is shrimp, 16% is crabs, 9% is scallops and so on. The table includes my own estimate of the mean live weight of each class of animals (my estimates are on the high side so as to generate a conservative answer to the question posed in the title of this post).
in the U.S. supply for consumption in 2010 to number of shellfish
|Class of shellfish||Percent by live weight||Live weight (in thousands of pounds)||Estimated mean weight per animal (in pounds)||Number of animals|
We eat over 40,400,534,000 shellfish each year. Using the formula mentioned earlier, we find that a vegetarian saves over 151 shellfish each year, the vast majority of them being shrimp.
How conservative are these numbers?
There is one reason why an economist may argue that these numbers are an overestimate. The existence of price elasticity of demand and the absence of price elasticity of supply in some sectors of animal agriculture suggests that a vegetarian saves fewer animals than he or she chooses not to eat. When vegetarians eat fewer animals, the price of meat drops and some people buy more meat. However, this argument from economics, I think, has a countervailing argument from sociology and psychology: a vegetarian often makes many other friends, family members, colleagues and fellow patrons at establishments eat more vegetarian meals or at least eat less meat than they would otherwise.
There are more reasons why the numbers derived in this post are conservative.
- This number does not include the hundreds of millions of fish “harvested” through recreational fishing at sea as well as in our rivers and creeks. This also does not include the number of land animals hunted and killed for meat.
- These numbers do not include bycatch, the fish we unintentionally catch and kill and then throw back into the sea. It is worth quoting Jonathan Safran Foer from Eating Animals here: “Imagine being served a plate of sushi. But this plate also holds all of the animals that were killed for your serving of sushi. The plate might have to be five feet across.”
- These numbers only include land animals who are counted by the USDA as having been slaughtered or condemned. They do not include species that the USDA does not count (e.g., rabbits and geese). They also do not include the fact that a larger number of wild animals die for animal agriculture than for vegetarian agriculture.
- At every point in the analysis above, I have always chosen the more conservative option, especially so in live weight estimates and conversions. In fact, the formula itself underestimates the number saved by a vegetarian because it only considers semi-vegetarians and Meatless Mondayers but not the full range of eating patterns.
A vegetarian, therefore, saves at least 406 animals per year (30 land animals, 225 fish and 151 shellfish).
My numbers are larger than the ones previously quoted in animal advocacy circles (usually 50, 95 or 100). Given how conservative my methodology is and how much larger my numbers are, vegetarians have been undercounting the number of animals they actually save and short-selling their impact on animal lives. Yet, this estimate is a work in progress. With more data becoming available and more meticulous tabulation of some things ignored in this post (such as bycatch), the estimated number is actually bound to increase.
Note: This post was updated on February 12, 2012 and on February 22, 2012.