Meat consumption patterns by race and gender

Do women consume less beef and pork than men? Do black Americans consume less dairy than white Americans? Answers to these types of questions are essential to developing targeted campaigns that employ some of the elementary principles of consumer marketing such as market selection and market segmentation. Dietary habits of an individual are complex products of that individual’s social and cultural context, influenced by a variety of factors including race, gender, ethnicity, age, income, social status and level of education. To successfully persuade large numbers of people to change their dietary habits with well-tailored campaigns, animal activists have to understand how food consumption patterns vary with these and other demographic characteristics.

Sources cited
  1. USDA, Economic Research Service. Commodity Consumption by Population Characteristics. (link, accessed January 1, 2013)
  2. USDA and USDHHS. Dietary Guidelines for Americans, 2010, 7th edition. December 2010. (link, accessed January 1, 2013)

Thankfully for us, for over five decades, the US Department of Agriculture (USDA) and the US Department of Health and Human Services (HHS) have collected data on individual food intake along with demographic data in extensive surveys. Since 2002, they have jointly administered an integrated program, known as the National Health and Nutrition Examination Survey (NHANES), which is uniquely comprehensive in that it includes both a physical examination of survey participants and a detailed interview involving a recall of everything consumed during the previous 24 hours. NHANES uses a representative sample of the US population and is widely used by scientists and medical researchers.

It is non-trivial, however, to convert the data from these surveys into an estimate of the amount of each commodity that enters the market for consumption per person. For example, a survey participant may report eating about two cups of cooked egg noodles but to disaggregate this data to estimate the amount of eggs that enter the market for it is not easy. Two divisions of the USDA (the Economic Research Service and the Agricultural Research Service) have collaborated together and completed exactly this task. They used data from the NHANES 2001-02 and other surveys, and first published their results in August 2011 and then updated them just last month (July 2012). The raw data used for this post may be found here.

By gender

Let’s begin with gender. The following shows food availability (which serves as a proxy for food consumption) in number of pounds in retail weight equivalent of each food category consumed per person per year separated by gender. Retail weight is the weight of a product as sold in the retail market. For example, the retail weight of a banana will likely include the weight of the peel. In case of meat, depending on how it is sold at the retail level, its retail weight may or may not include the weight of bone, fat, or inedible trimmings.

Meat consumption by gender in the United States
(in retail weight pounds per person per year)

The data show that women eat substantially less meat than men, especially beef and pork. Compared to men, however, a larger proportion of the meat consumed by women is chicken or fish. A question that arises, of course, is if women over time have been replacing beef and pork in their diet with chicken and fish. This is of concern to animal activists who care about individual animals and not just the weight of meat consumed. They value the interests of a five-pound chicken as much as that of a thousand-pound steer and so, a shift from beef/pork to chicken/fish causes harm to more animals (e.g., see Vegan Outreach make this argument here). Unfortunately, this data is not enough to definitively answer the question—for that we will need more of a longitudinal study with repeated surveys over an extended period of time.

But there are some related questions that the data can answer. To gain a complete picture of the relative consumption patterns of men and women, take a look at the figure below which illustrates consumption data for each major food category.

Food consumption by gender in the United States
(in retail weight pounds per person per year)

If we look at the per capita consumption of all major categories, we find that women, on average, eat about 20% less than what men eat (in retail weight). This is crudely consistent with the difference in healthy body weights of men and women and also with the dietary guidelines released by the USDA and the HSS which suggest that caloric needs of women are about 20% less than those of men. But women eat 42% less beef/pork, 23% less poultry and 21% less fish than what men eat (all a larger reduction than the average of 20%). This is compensated by the fact that women eat only 10% less fruits and only 18% less vegetables than what men eat. In fact, if we dig even deeper into the data, we find that women eat more of the healthiest vegetables than men. Women eat 19% more cruciferous vegetables, 12% more broccoli and 14% more leafy greens than men (not illustrated in the bar graphs in this post). So, on average, in relation to men, women are compensating for their proportionally lower intake of beef/pork with a proportionally higher intake of fruits and vegetables, especially the most healthful vegetables! But, it is perfectly possible that both men and women are replacing beef/pork with chicken/fish and this data cannot conclude one way or the other.

That women eat only 65% of the meat that men eat leads to some additional inferences. For example, on average and assuming no other influencing factors, persuading a man to give up meat only on weekdays has about same impact on animals as persuading a woman to commit to becoming vegetarian. This, however, does not imply that vegetarian outreach is more effective when targeted at men rather than women (or vice-versa) because such effectiveness can depend on other influencing factors such as differences between men and women on persuadability and openness to a change in diet.

By race or ethnicity

The following bar graph illustrates the different categories of meat consumed by United States residents, separated by race or ethnicity. As in earlier graphs, the numbers shown are in pounds of retail weight consumed per person per year.

Meat consumption by race or ethnicity in the United States
(in retail weight pounds per person per year)

The data show that Hispanic Americans consume more beef than other racial or ethnic groups. Black Americans consume a significantly larger amount of chicken (at least 38% more), turkey (at least 38% more) and fish (at least 53% more) than any other racial or ethnic group included in this data. Given that consumption of chicken and fish causes harm to a significantly larger number of animals than the same amount of consumption of beef or pork, this data should inspire animal activists of a utilitarian persuasion to intensify their efforts to reach black Americans with their message.

By the way, for the sake of completeness, here is the per capita consumption data on all food categories separated by race or ethnicity.

Food consumption by race or ethnicity in the United States
(in retail weight pounds per person per year)

On a final note, demographic characteristics such as race or gender can be extremely useful in building effective outreach strategies but can also trick our minds into engaging in unwelcome stereotyping. Individual behavior within any demographic group varies greatly; we stereotype when we judge an individual solely by his or her demographic group. In using this data, let’s heed the advice frequently given in consumer marketing courses: segment, but do not stereotype.

Note: This post was updated on August 24, 2012.

Comments

Nick

I have a bunch of thoughts about your post! First, AWESOME to see you covering this. I've been looking at some of the same data (both from separate studies, and also one study that looked at the NHANES data) and thinking about the same thing, and great to see you lay it out in such a clear way.

There's one important consideration though to keep in mind when deciding which group to target, which isn't mentioned in the blog post, which is likelihood of change. So for the men/women question, leaving aside beef and pork (since as you point out they virtually don't matter in terms of helping animals), men eat 20% more chicken and fish (so maybe 22% more animals) than women. But women are almost twice as likely as men to go vegetarian, and also make more of the food purchasing decisions in households than men (I can't find the exact percentage). The latter is clearly changing, though still relevant, but the former doesn't seem to have changed much over the past 15 years.

So, it's still better to focus on women if targetting by gender. In theory the same outreach would net 100 men or 200 women going veg, meaning (even leaving aside purchasing power, which is important), meaning about 56% more animals spared (assuming my math here is right: 100 men x 1, 200 women x .78) if focusing on women. Now it's certainly possible/likely that the success rate of outreach does not 1:1 mirror the current vegetarianism rates. (Men may be more close to women in ability to be persuaded, and less likely to go veg on their own; or it could be the reverse). But, until we know any other info on that or strong reasons to think men are as or almost as likely to be persuaded, I'd assume it's safest to go with the 1:1 ratio. (My personal experience leafleting and doing humane ed would suggest women are much more likely to be persuaded just as they are much more likely to go veg.)

Harish

Nick, you are absolutely right to note that I did not bring up persuadability in the post (on purpose). While it is very likely true that women are more persuadable than men with regard to vegetarianism, I did not find any good studies on it based on American residents or hard data of my own (other than the fact that more women are vegetarians). That is why when I compared advocacy for men vs. women, I used the escape clause "assuming no other influencing factors".

Anyway, my post does not take a position on whether it is better to focus on men or women because I wanted to stick to only conclusions that I could justify with numbers in published studies. But, it is possible that the post gives the impression that it is suggesting a focus on men for veg advocacy. So, I added a sentence at the end of the gender section: "This, however, does not imply that vegetarian outreach is more effective when targeted at men rather than women (or vice-versa) because such effectiveness can depend on other influencing factors such as differences between men and women on persuadability and openness to a change in diet."

Nick

One of the issues you raise is the troubling issue of people (women specifically, though it looks like men are doing this more recently as well) switching from beef to chicken. You're suggesting that women are replacing beef/pork with more fruits and vegetables, not chicken, which makes sense but I don't totally understand how you can conclude that from the data. Can you explain to me how we can infer that women aren't eating both more chicken/fish AND more fruits/vegetables?

Harish

Women may well be eating both more chicken/fish AND more fruits/vegetables, but only if men were likely doing so too. I think I did not make completely clear in the post that I was only looking at women's consumption relative to men. I was only suggesting that women, in relation to men, were not eating proportionally more chicken/fish to compensate for eating proportionally less beef/pork (assuming also that women have always been eating about 20% less than men). But, it is perfectly possible that both men and women are replacing beef/pork with chicken/fish and this data cannot conclude one way or the other. Thanks for the question. I edited the paragraph a little to clarify this point.

Spike

I am pleasantly surprised to see that the fruit/vegetable/grain numbers together exceed the animal product numbers. I guess the pessimistic picture I sometimes have in my head of the average American living on a diet of meat and cheese isn't true. Of course, anything above 0 in the animal product categories is depressing, but having this data is helpful in thinking about new targeted strategies.

Harish

Spike, an interesting limitation of this data is that it is by retail weight. For example, it is not possible to compare a retail weight pound of bananas with a retail weight pound of blueberries because the former will include the weight of the peel which is not eaten. In general, in this data, we cannot readily compare the meat eaten by men to vegetables eaten by men, but it is reasonable to compare the meat eaten by men to meat eaten by women. For example, the weight of fruits may be inflated because of fruits like watermelons with heavy inedible portions and so it cannot be compared to the retail weight consumed of, say, dairy.

This is why I have not drawn any pie charts in this post because different commodity groups cannot be compared against each other. But, a commodity eaten by one demographic group can be compared against the same commodity eaten by another demographic group.

By the way, just so you don't get too encouraged about the American diet, almost a third of the vegetable portion is potatoes (which, I suspect, is mostly in the form of potato chips and french fries).

Paul S

This is a fascinating and very worthwhile post (and comments!). Thanks so much for doing this research, Harish. Your blogs are always so insightful.

Harish

Thank you, Paul, for your very kind words!

Matt

Great stuff as always, Harish! I would be interested in any data by age, too. Do HS / college age kids eat more red meat or fish than the general population? Etc. Our daughter's friends never eat fish, for example.

Harish

Thank you, Matt, for your comment and question. Prompted by your question, I looked into the data and you are absolutely right about teenagers not eating fish!

Boys (age group 12-19) eat about 85% of the meat that men (age 20 or above) eat, but boys eat only 30% of the fish that men eat. Similarly, girls (age 12-19) eat about 89% of the meat that women eat, but only 46% of the fish that women eat.

Boys eat the same amount of beef as men do, and girls eat about the same amount of beef as women do.

So, yes, fish is not very popular among young people! Your daughter's experience, I would guess, is not atypical.

But, unfortunately, chicken seems quite popular among them. According to this data, girls eat more chicken than women do!

Spike

I know your blog numbers are based on "consumption", but I wonder if it's the case that when women report consuming more broccoli it is actually true that they buy more but then throw it away. I was wondering why so many people are so overweight and unhealthy if they are consuming so many fruits and vegetables.

According to this report (link to cnn article below), "American families throw out approximately 25% of the food and beverages they buy."

http://news.blogs.cnn.com/2012/08/22/40-of-u-s-food-wasted-report-says/

Harish

Spike, you are right that the numbers in this post are based not on what is actually eaten but on what enters the food market (what USDA calls "food availability"). Food availability serves as a proxy for food consumption, but they are certainly not the same. But, animals activists care mainly about food availability because a hen suffers to produce an egg whether or not the egg is eaten.

The USDA has not completely described its methodology behind these numbers for us to know how they account for waste in a consumer's home, on the restaurant plate, or at the grocery store. But, they do talk about these types of "losses" and consider them.

So, we don't know if people are throwing away a larger fraction of the broccoli they buy than the meat they buy. If that is the case, it would explain some of why people are unhealthy despite such high "consumption" of fruits and vegetables.

But, let's remember that a third to a half of vegetables people eat is potatoes (e.g., as chips and fries) and corn (e.g., as popcorn). A huge chunk of the fruit Americans eat is not as whole fruits but in the form of juices, jellies and jams with tons of added sugar. So, a large retail weight of fruits and vegetables can still mean a less-than-optimal diet.

Spike

Thanks for clarifying! I guess it's all around bad news: People are eating a lot of fruits and vegetables in the form of junk and they are buying food and throwing it away! Buying animal products and throwing them away is particularly sad.

Dave C-H

Terrific post. This is quickly becoming my favorite blog.

Harish

Thank you, Dave! Glad to hear so.

Jon

Fascinating and very useful for our outreach efforts. Thanks, Harish!

Matt

Thanks, Harish. Great post / comments.

Katrina

Thank you for making this information available!

Sarah

Spike- Thanks for being the one sexist douche bag to post something! Are you upset that the data doesn't favor men?

Please add your comment below. Both of the two fields below, your name and your comment, are required.