List of pages in this section
- The most consumed vegetables in the USA
- The most consumed fruits in the USA
- The most consumed legumes in the USA
- The most consumed grain products in the USA
- The most consumed meat, poultry and fish in the USA
- The most consumed dairy products in the USA
- The most consumed added fats in the USA
- The most consumed nuts in the USA
- The most consumed sweeteners in the USA
While researching for blog posts, I often need impartial statistics about which foods are the most popular or important in the American diet. I found it difficult to find this information from trustworthy sources, so I decided to compile the data myself. Hopefully, other people researching the topic can use this data and avoid frustration.
The most complete source for this type of data I’m aware of is from the Economic Research Service of the United States Department of Agriculture. They maintain the Food Availability (Per Capita) Data System, which records the national availability of a wide range of foods per-person in the USA, going back to the 1970s.
The data are freely available in XLS (Excel) format. However, the formatting of the data can make it difficult and laborious to rank and compare the availability of different foods. It can also be cumbersome to view XLS data in-browser. I’ve taken data from these tables and rearranged them to make them more accessible and to make comparison between foods much easier.
I’ve also added calorie information from the Agricultural Research Service National Agricultural Library. Calorie information is included per hundred grams, and is also computed to show the availability of each food in calories per person. The idea behind including this information is to provide a measure of how important each food as a provider of food energy — I think this perspective can be valuable when comparing pounds of potatoes with pounds of lettuce.
Calorie information is based on raw food in the least refined state in which it is commonly sold. When dealing with fresh-weight-equivalency (e.g. for fruit juice), this information is approximate but should be a serviceable estimate.
Where the availability data encompasses a number of distinct foods under a single term (e.g. for fruits, the “melons” category includes cantaloupe, watermelon, honeydew etc.) calorie information for the most commonly eaten variety is used.
What food availability means
“Availability” measures how much of a certain food is obtainable for human consumption in any given year. The ERS describes it as a “popular proxy” for how much of a given food is consumed in America.
It’s a “proxy” in the sense that it doesn’t directly measure the statistic we’re interested in — food consumption. But by measuring how much food is produced and available to be eaten, we can use food availability as a rough approximation for food consumption.
But why not just measure food consumption directly? Chiefly because it’s extremely complex and expensive to do so well. Food consumption patterns vary widely across cities, states and regions, so it would be imperative to sample thousands or tens of thousands of people from all over the country to ensure the data collected were representative of the nation as a whole. Moreover, getting people to accurately record everything they eat is extremely difficult, especially when dealing with ready-to-eat food purchased from restaurants or stores. How many ounces of onion were on that burger? How many carrots were in that vegetable soup? How much garlic? How much pepper?
The approach of the food availability system is to measure how many onions or carrots are grown, prepared and available in stores or wholesalers for human consumption. This is less direct than measuring what people actually eat, but it’s much simpler and cheaper to keep track of, because most of these data are recorded by farmers, processors and retailers anyway.
The basic formula for availability of a given food is: (domestic production + imports + initial stocks) – (agricultural use + exports + ending stocks)
For example, to measure the availability of onions, you would first measure the sum of all onions harvested that year, all onions imported into the country and all existing stocks of onions ready for sale. You would then subtract all onions used in agriculture (e.g. used in animal feed or used to produce more onions), all onions exported to other countries and all onions left in stock at the end of the year. The difference between these numbers would be the total availability of onions that year. To get availability per capita, you would simple divide this number by the number of people living in the USA.
For many foods, the data are also adjusted to account for how the food is processed and distributed. For example, how meat is boned and trimmed, or how vegetables have inedible portions removed.
Limits of the data
The first obvious limit is that availability is not the same as consumption. Not all food that is available to be eaten is actually eaten. For this reason, it’s important to not interpret lb of carrots available per person as lb of carrots eaten per person. The primary use of these figures it to make comparisons between foods. For example, you might reasonably assume that if there is twice as much onion available per person than carrot, consumption of onion is about twice as much as of carrot, even if you can’t tell the absolute value of consumption for either vegetable.
In order for this to be a valid comparison, the ratio between availability and consumption has to be roughly equal between each food in a given category. This will not always be the case because the factors influencing this ratio may be different for each food.
Much of the difference between availability and consumption is due to wastage and spoilage of food in the supply chain or in people’s homes. The speed at which a food spoils will increase the chance that has to be thrown out before it can be eaten. Therefore, we might reasonably expect a food which spoils faster to have a bigger difference between availability and consumption than a food which can be stored for a long time.
In the vegetable category, for instance, carrots have a relatively long shelf life while lettuce has a relatively short shelf life. The ratio of availability to consumption might be expected to be different for these two foods, but the tables presented here will not account for that.
Another potential source of error is the proportion of a given food typically discarded during preparation. While the data are adjusted to an extent to take this into account at the farm and processing level, they do not take it into account at the store, restaurant or domestic level. Accordingly, a vegetable like broccoli, the preparation of which typically involves discarding a large portion of the edible stalk, might be over-represented in these tables. When cooking oils are used for deep-fat-frying, only a small portion of the oil may actually be consumed.
Data use and licensing
The data used in these tables are taken from public sources. They may be copied and distributed with no restrictions other than those imposed by the primary sources. If you do use these data in a project or publication, a link back to this page is requested.
The ERS Food Availability Documentation should be considered essential reading for thoroughly understanding the food availability data.
Special thanks to Jeanine Bentley of the USDA ERS for very patiently answering some of my questions about food availability data. Neither Ms. Bentley nor anyone else from the ERS has reviewed this section of the website; any errors or misinterpretations are entirely my own.