Meat consumption and demand both in decline

We have known for sometime that per capita meat consumption in the US has been on a steady decline. But, consumption is not the same as demand. For example, when prices rise or incomes fall, the per capita consumption of a product may drop but not necessarily the demand for the product. Animal advocates are interested in reducing per capita meat consumption, but they are even more keen on actually reducing the demand for meat. Meanwhile, the meat industry blames the decline on everything but a slump in demand. They have a point—after all, driven by several factors including high feed costs, the retail price of all categories of meat have increased substantially during the last few years and it suggests an explanation other than reduced demand for the reduced per capita meat consumption.

In this post, I examine if it is not just the per capita meat consumption but the demand itself that is also in decline. Let’s begin first with a look at the consumption numbers.

Per capita meat consumption to continue decline through 2013

The USDA, on November 16, released its most recent forecast of meat production and consumption through the rest of 2012 and the year 2013. These projections indicate a continued decline in the per capita consumption of all four major meat categories: beef, pork, chicken and turkey. The graph below plots the per capita consumption (as implied by per capita disappearance of a product into the market) of meat products beginning with 1966 (the first year for which per capita disappearance data is available for broiler chickens in the food availability data released by the USDA.)

Per capita consumption of meat products by retail weight in pounds (except turkeys, which are reported by carcass weight)

Per capita chicken consumption is projected to drop by more than 8% in 2013 compared to the peak reached in 2006. During this same period, per capita beef consumption is projected to drop by about 17%. The per capita pork consumption is projected to drop by almost 11% since a recent peak in 2007 and the per capita turkey consumption is projected to drop by more than 9% since a recent peak in 2008.

Nice. But is the demand for meat also in decline?

The decline in per capita consumption does not necessarily imply a decline in demand. For any product, economists usually relate price and demand by considering a measure called the compensated demand elasticity. This measure for any product incorporates factors such as the share of income spent on the product, the percentage change in the price of the product and the percentage change in the per capita consumption of the product.

Sources cited
  1. USDA, Economic Research Service. Livestock, Dairy, and Poultry Outlook (multiple reports from 2006 to 2012). (link, accessed November 20, 2012)
  2. USDA, Economic Research Service. Food Availability (Per Capita) Data System. August 20, 2012. (link, accessed November 20, 2012)
  3. G. T. Tonsor, J. R. Mintert, and T. C. Schroeder. U.S. Meat Demand: Household Dynamics and Media Information Impacts. Journal of Agricultural and Resource Economics 35(1), 2010. (link, accessed November 20, 2012)
  4. USDA, Economic Research Service. Meat Price Spreads. November 16, 2012. (link, accessed November 20, 2012)

The compensated demand elasticity is typically a negative fraction. For example, if it is −0.3 for a product, it means that if the price of the product increased by 1%, one should expect that its per capita consumption would reduce by 0.3%. It is a compensated metric because it aims to remove effects of income in this calculation. For example, it will correctly predict that the per-capita consumption of a cheap product (relative to income) like table salt will not change much with a change in its price.

For each of the four categories of meat, the table below lists the compensated demand elasticity reported in a paper published in 2010 in the Journal of Agricultural and Resource Economics. This computation is based on quarterly USDA data from 1982 to 2007. The table lists the price in 2006, the corresponding inflation-adjusted price in 2006 (using the CPI Inflation Calculator provided by the Bureau of Labor Statistics) and the actual price in 2012 (my source for prices include USDA data on meat price spreads and the monthly USDA reports on Livestock, Dairy and Poultry Outlook.) Prices are in cents per pound by retail weight for each of the meat categories. The percentage change in deflated retail price multiplied by the compensated demand elasticity gives us the per capita consumption one would expect in the absence of a change in demand (listed in the second to last column). In the last column, the table includes the actual percentage change in per capita meat consumption during the same period from 2006 to 2012.

Changes in price, per capita consumption and implied demand for each category of meat between 2006 and 2012 (prices are in cents per pound of retail weight)
Price in 2006 Inflation-adjusted price in 2012 cents Price in 2012 Percent change in deflated price Compensated demand elasticity Expected change in per capita consumption Actual change in per capita consumption
Beef 397.02 455.54 500.04 9.77% −0.420 −4.1% −13.6%
Pork 280.80 322.19 346.95 7.68% −0.740 −5.7% −6.7%
Chicken 157.07 180.22 188.04 4.34% −0.099 −0.4% −6.6%
Turkey 77.00 88.35 105.80 19.75% −0.099 −2.0% −3.6%

Take a look at the last two columns and note the larger percentage change in the last column in comparison to the last but one. The actual per capita meat consumption has declined far more than what can be attributed to changes in the price of meat brought about by any of the reasons usually given including reduced production, higher feed costs and drought.

Nearly 70% of the decline in per capita consumption of beef since 2006 is likely due to a decline in demand. More than 93% of the decline in per capita consumption of chicken is also likely due to a decline in demand.

Dear animal advocates, rejoice this holiday season in the knowledge that the demand for meat—and not just consumption—is on the decline!

Postscript (7/10/2014): The forecasts made by the USDA in November 2012 about per capita consumption in 2013 turned out to be an underestimate. Per capita consumption in 2013 of beef, pork, chicken and turkey was 56.4 pounds, 46.8 pounds, 81.9 pounds and 16.0 pounds, respectively. Only per capita beef consumption continued its decline in 2013. Year-to-year fluctuations are to be expected—it is possible but still too soon to infer that the declining trend has stopped entirely.

Comments

Matt Ball

You rock, Harish! Great work on everyone's part to bring about the awareness behind this decline!

Paul Shapiro

I echo Matt's comments. Thanks so much for this insightful--and promising--post, Harish! The animal movement keeps on winning.

David Coman-Hidy

Another amazing post! Always nice to see some hopeful news to remind us that real change is possible!

Erik Marcus

This is so incredibly good. It's wonderful to see such serious scholarship and hard numbers on this vitally important issue!

Mahi Klosterhalfen

Great work, Harish! And congrats to my colleagues in the US who make this happen!

great stuff

thank you so much, people are waking up.. it's happening.

Kelly

Great job! Reason for hope!

John T. Maher

China's consumption is rising faster than US consumption is declining

Laura Hart

This is an excellent and uplifting piece. GREAT news-- very uplifting!! Thank you!

Jon Camp

Love it. Thanks, Harish!

Jack Norris

Really great news - thanks for figuring this out, Harish!

Ben Davidow

Great news and great analysis, thanks for posting this!

Mary Prubant

This is wonderful news. Great story. Thank you.

Debby Sunshine

Job well done! Reason to rejoice! I like your use of graphs!

Robert Davy

Thank you for posting this insightful analysis. In this blog post I took a look at the global situation using (somewhat older) FAO data up to 2009.
http://canberravegan.blogspot.com.au/2012/11/world-trends-in-consumption.html

Lewis Bollard

Thanks for the great analysis Harish!

Steve Erlsten

Thank you for crunching the numbers! The impact of rising prices is in the back of my mind every time I talk about Americans' reduced meat consumption, and this backs up my hunch that actual demand is dropping!

Mikael Nielsen

This is great and I will share far and wide. Thank you!!

Bea Elliott

Sadly I have to agree with John T. Maher
that China's consumption is rising faster than US consumption is declining.

But on another positive note (sort of) we can see that animal ag is desperate to create and maintain "demand" for their meat by telling the Obama Administration that they MUST include agriculture in trade deals. Calling the EU "unreasonable", "unscientific" and "arbitrary" if they spurn US meat. I often wish there wasn't "free trade".

http://nationalhogfarmer.com/business/us-trade-deal-eu-must-include-agriculture

RespuestasVeganas.Org

Good news!! :)

Zia Terhune

Are dairy and eggs declining as well?

Harish

Zia, fluid milk consumption is declining but not necessarily all dairy products (especially cheese, the consumption of which is increasing somewhat). Egg consumption is more-or-less declining. These are, I am hoping, topics for future posts.

Ben West

Awesome post, Harish! One question: the elasticity of meat will presumably change with income - do you think lower incomes as a result of the recession could be behind some of the decrease in demand?

Harish

Ben, you are absolutely right that change in income can change demand just as well as a change in price can. In the case of meat, however, evidence suggests that reduced incomes due to the recession are unlikely to be a significant reason for the reduced demand for meat.

Economists measure this relationship between income and demand using a different metric called the income elasticity of demand (as opposed to the price elasticity of demand). The studies that I have looked into suggest that the income elasticity of beef is close to zero and the income elasticities of other meats are negative. In fact, a recent (2010) meta-analysis of the income elasticity of meat, published in the Australian Journal of Agricultural and Resource Economics, confirms this. Higher income is correlated with reduced consumption of pork and poultry and more or less unchanged consumption in beef and fish. Some other studies based in the US have estimated even beef to have a negative income elasticity. This means that increased income does not actually increase consumption, or reduced income does not actually reduce consumption of these products (although changes in price may change consumption assuming no change in income). This is vaguely consistent with the fact that, in the richer countries like the US, higher income is correlated with increased consumption of fruits and vegetables.

So, my guess is that lower incomes as a result of the recession are not a significant factor behind the decline in demand for meat in the US. But, we will know for sure only when the economy recovers fully.

James

A problem is that the relevant information for animal advocates isn't "price versus non-price factors", it's "persistent versus impersistent factors".

Most, but not all (e.g. tax), price factors are also impersistent so removing it helps. However, a lot of non-price factors are also impersistent.

E.g. For the animal advocate, there's no relevant difference between a temporary fall in meat consumption because of a short burst of negative media attention and a temporary fall in meat consumption because of price volatility. However, there is a difference between both of these cases and a decrease in meat consumption because of an increasing number of veg*ns.

Caryn Ginsberg, Animal Impact

Super analysis, Harish!

Question on the income elasticity studies that show no or even negative relationships between income and consumption. The research seems to look cross-sectionally at richer vs. poorer individuals (or countries). Maybe it makes sense that meat consumption goes down due to better education about and/or access to fruits, veg, grain, etc..

But the recession would appear to be a different effect, reducing incomes for individuals over time in a way that didn't necessarily change their core preferences. If someone loses a job, they didn't suddenly get more educated on health, fruit and veg, etc. Thoughts?

I've wondered how much of the decline in per capita meat is, in fact, a budget issue in tough times. Good news is that the decline in chicken and beef that you show starting in 2007 would predate the recession.

Thanks for this great piece.

Harish

Caryn, As you say I suspect that part of the reason income elasticity of meat is negative is that higher income is correlated with better nutrition education and access to fruits and vegetables.

But, I think the other and bigger reason is simply that fresh fruits and vegetables cost more than processed meat (a head of broccoli costs twice as much as a McDouble with two beef patties or a McChicken). So, when someone loses a job, they would save more money eating at McDonald's than eating fresh fruits and vegetables. I am not sure but I think this is a bigger reason why published studies show that the income elasticity of demand for meat tends to be zero or negative.

So, that is why I think reduced income due to recession or tough times is not likely a substantive factor in this decline of meat consumption. As you note, some of the current decline predates the recession. Also, there was no decline outside of the larger pattern of changes during the 1991-92 recession either.

While I think reduced income may not be a reason for the decline, price still is a factor. When prices rise above what people are used to paying, more people think it as overpriced and choose not to buy the item. For example, you are likely to balk at paying $10 per pencil even if you can afford to do so.

ardeth

I'm wondering if the USDA took into account the current popularity of so-called free-range "organic" eggs and humanely raised meat animals from small farms. Is that included in the graphic information, Harish?

Harish

ardeth, This post does not concern eggs. Regarding meat, the USDA food availability data is primarily based on production estimates made by the National Agricultural Statistics Service (NASS). According to the NASS, these estimates are based on a survey of "all known farms and ranches". Small farms are included in the list that forms the basis of the survey. Likely because of the large number of small farms, NASS says it uses a random sampling methodology in the survey of smaller farms.

So, yes, small farms are included in the graphical data.

neuran

Harish

neuran, This post is about US meat consumption and demand, not global meat consumption. WWI is not saying anything contrary.

Guy Grayson

Harish your innovation and intelligence are just what the animal rights movement needs- we can all learn a lot from your example. When intellect meets purpose the animals have a fighting chance!

Cathie

Harish thank you for putting this together. I am encouraged. Hoping for better and better results for the animals like we all are.

Bob Unferth

This is an interesting analysis. But you recognize its fundamental weakness:
the analysis assumes that the indicated compensated demand elasticities are
accurate - with only an appeal to the authority of a single paper published in
the Journal of Agricultural and Resource Economics. It is the crucial issue,
passed off with an appeal to authority.

Consider the reported compensated demand elasticity for chicken. Purportedly an
increase in the price of chicken from $1.80 per pound to $1.98 (10%) would yield a 1% decline in consumption. Or a drop in price from $1.80 to $1.62 would yield a 1% increase in consumption. Does that seem realistic? Seems pretty low to me.

Try a more extreme example: say the price of chicken dropped in half, would
demand rise by only 5%? Really?

A couple of caveats concerned such thought experiments:

1. We intuit only the short term impact of these price changes, but the actual
concern is with the long term impacts.
2. Elasticities change with different price points - people may respond
differently to a 10% increase in the price of chicken from 90 cents per pound to 99 cents per pound than from a 10% increase from $3 per pound to $3.30.

But, overall I find the reported elasticities for chicken and turkey a bit implausible.

A second reason to question the reported compensated demand elasticities is the
differences. Why should a 10% increase in the price of pork result in a 7½% (or
4% in the case of beef) decrease in demand, while a 10% increase in chicken or turkey result in only a 1% decrease in demand? Why should the elasticities
differ by factors of 7½ and 4?

A third reason to question the reported compensated demand elasticities is the
effect of declining median incomes. From 2006 to 2011 median income in the US fell by about 7%. The effect of falling income on these elasticities is not mentioned, but is obviously of the greatest significance. A fall in income of 7% means that disposable income - what's left after paying the rent, etc, is reduced by a much larger percentage.

A last problem with this analysis: It depends on estimates of prices, inflation and consumption for the year 2012 (the date on the paper is November 20) and forecasts of consumption through 2013. John Kenneth Galbraith had this to say: "The only function of economic forecasting is to make astrology look
respectable."

In conclusion, more study is needed.

But my nonsense and wishful thinking detectors are quivering.

Sorry to rain on everyone's parade.

Harish

Bob,

Thank you for taking the time to write a detailed comment!

Food is very inelastic because everyone needs food no matter what the price. Chicken being among the cheapest of what people eat, it is very reasonable that the elasticity of chicken is at 0.099. That is why it is very reasonable and realistic that when the price of chicken rises from $1.80 to $1.98 per pound, there will be only a 1% decline in consumption. As much as I wish, people do not change their eating habits much over a dime or too. If you can point to a flaw in the paper that computes this elasticity based on a quarter-century of data, then maybe you would have a point. Until then, hard data should be our guide instead of any one individual's intuition (yours or mine).

It is also important to remember that the elasticity number usually applies only for small delta changes, not arbitrarily large changes in price. And, they are different for different price points as you also point out. Elasticity numbers are computed assuming current prices compared to income and reasonable price changes. So, using extreme examples does not demonstrate that the 0.099 number is wrong.

Another reason for your skepticism is that the elasticities are different for different meats. But, they have always been different (from studies dating back to the 70s) for complex cultural reasons which give different levels of emphasis to different meats in diets.

Regarding why impact of income is not mentioned, see the discussion above on income elasticity. In richer countries like the US, most meats actually have a negative income elasticity! Declining incomes do not even enter the picture as a factor in reduced meat consumption. A meal of KFC chicken is often a cheaper meal than one with fruits and vegetables.

You are raising the issue that the elasticity numbers are based on just one paper. Your point would be a better one if you pointed to a flaw in the paper or if you pointed to other papers that made similar computations and arrived at different numbers for elasticity.

Your last problem with the analysis is that you think it depends on forecasts of consumption through 2013. But, in fact, it does not. I apply the elasticity numbers only up to 2012 to the actual drop in consumption, not the forecasted drop in consumption. The 2013 forecasts come from the USDA.

The parade is still on. It will be at least a few years before we know for sure if this parade is justified.

howard Greenebaum

The comparison of a head of brocoli with a single sandwich does not make mathematical sense. A head of brocoli will be a source of several meals.
Does anyone know of any comparisons of the cost of groceries of a meat eating family and a vegan eating family? This comparison is very important to a project I am working on. Please send info to my email address. thank you.

Samuel

Is there similar data on marine animals?

Harish

Samuel, USDA livestock reports do not include fish and I do not know of 2013 projections for fish consumption. But, according to the food availability reports from the USDA, fish consumption also fell by about 5% from 2006 to 2010.

Nancy Wheeler

Harish: Great news and great analysis. As as vegetarian - almost a vegan - I am delighted. I do not believe in eating any anbimals and that includes marine animals. Any hopes for them? How can we promote less eating of marine life?

Spike

Great news! Love it!

Jack Halperin

Consumption of chicken and turkey have not gone down throughout most of the period.

Bettin-sky

Yeah it's nice the prices would help us

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