Carpe Diem

Something to be thankful for: Real cost of a Thanksgiving dinner for ten this year is 1.4% cheaper than last year

From the American Farm Bureau Federation (AFBF):

The retail cost of menu items for a classic Thanksgiving dinner including turkey, stuffing, cranberries, pumpkin pie and all the basic trimmings increased less than 1 percent this year, according to the American Farm Bureau Federation (AFBF).

AFBF’s 27th annual informal price survey of classic items found on the Thanksgiving Day dinner table indicates the average cost of this year’s feast for 10 is $49.48, a 28-cent price increase from last year’s average of $49.20.

“At just under $5 per person, the cost of this year’s meal remains a bargain,” said AFBF President Bob Stallman, a rice and cattle producer from Texas. “Our diverse farm and ranch families are honored to produce the food from our nation’s land for family Thanksgiving celebrations. During this holiday season, I am encouraging farmers and ranchers to reach out to consumers in-person or through social media, to answer questions about the food that they grow or the livestock and poultry they raise.”

The AFBF survey shopping list includes turkey, bread stuffing, sweet potatoes, rolls with butter, peas, cranberries, a relish tray of carrots and celery, pumpkin pie with whipped cream, and beverages of coffee and milk, all in quantities sufficient to serve a family of 10. There is also plenty for leftovers.

The big ticket item – a 16-pound turkey – came in at $22.23 this year. That was roughly $1.39 per pound, an increase of about 4 cents per pound, or a total of 66 cents per whole turkey, compared to 2011. The whole bird was the biggest contributor to the final total, showing the largest price increase compared to last year.

“Thanksgiving Dinner is a special meal that people look forward to all year,” said John Anderson, AFBF’s deputy chief economist. “Most Americans will pay about the same as last year at the grocery store for a turkey and all the trimmings. A slight increase in demand for turkey is responsible for the moderate price increase our shoppers reported for the bird,” he said.

Some comments:

1. Compared to last year’s cost of $49.20 for a complete classic Thanksgiving dinner, this year’s cost is only 0.57% (and 24 cents) higher at $49.48 (see red line in chart).  Even though turkey went up this year by 3% compared to last year, almost all of the other items decreased in price since 2011 by an average of more than 2%: whipping cream (-6.6%), stuffing (-3.8%), potatoes (-3.4%), milk (-1.9%), cranberries (-1.2%), peas (-1.2%), pumpkin pie mix (-0.33%) and pie shells (-0.40%).

2. Adjusted for inflation, the cost of a classic Thanksgiving dinner for ten this year is 1.40% cheaper than last year (see blue line in chart).

3. Compared to the inflation-adjusted cost of more than $60 in 1986, today’s classic turkey dinner for ten is 17.6% cheaper.

4. Measured in time worked at the average hourly wage for all private workers of $23.58 in October 2012, the “time cost” of this year’s classic turkey dinner for ten is only 2.10 hours, down slightly from 2.12 hours last year.  

5. Cost conscious shoppers can buy the same classic Thanksgiving meal at Walmart for only $36.72 (see top chart above), a savings of more than 25 percent compared to the AFBF national average, according to this press release from Walmart. In hours of time worked at the average hourly wage, that would be a “time cost “of only 1.56 hours for a holiday feast for ten at Walmart, an amazing bargain.

6. It’s not just the cost of the classic Thanksgiving dinner that has remained so affordable over time, spending on food for Americans in general has gotten more and more affordable over time, relative to our income.  The bottom chart above shows that total spending on all food consumed at home in 2012 will be less than 7 percent of the total disposable income for Americans. In comparison, Americans in 1970 spent twice as much of their disposable income  on food (14%), and in 1950 Americans spent almost three times as much on food at home as a share of income (19.7%).

Bottom Line: The fact that the average American family can celebrate Thanksgiving with a classic turkey feast for ten people for less than $50 and at a “time cost” of only 2.1 hours of work for one person (and only 1.56 hours for Walmart shoppers) means that we really have a lot to be thankful for on Thanksgiving: an abundance of cheap, affordable food.  Relative to our income, food has been more affordable in recent years than at any time in history.

Bon appetit!

Update: Here are USDA data on food expenditures, they are reporting in Table 7 that spending on food at home as a share of disposable personal income was only 5.7% in 2011. Keep in mind that the calculation is based on total expenditures on food at home ($660 billion) as a share of total personal disposable income ($11,593 billion).

21 thoughts on “Something to be thankful for: Real cost of a Thanksgiving dinner for ten this year is 1.4% cheaper than last year

  1. I’m at 17.85% of my income for food from grocery stores averaged over the 11 months of this year (I use YNAB budget software). We don’t eat steak or lobster, and there is just two of us.

    The BEA percentage seems low for everyone I know. Here’s an example I popped into an Excel spreadsheet. I considered disposable pay as take-home pay, these are just estimates, and I interpreted 7% in 2012 from the BEA chart above. I don’t know anyone who only spends $50 per week on groceries.

    $50,000.00 Gross pay per year
    $37,500.00 75% take-home pay per year
    $2,625.00 Groceries at 7% per year of take-home pay

    $50.48 Groceries per week

  2. this seems like a very bad metric for “cheaper”.

    the price went up.

    if the price of cars or whatever other things went up more, that does not make it cheaper. it’s just relatively less additionally expensive than other items.

    if you want to talk about affordability (which seems a better term here than cheapness) then it would seem you would want to compare it to income or perhaps disposable income.

    merely saying the price of dinner is up less that the price of sneakers does not really say anything about either cheapness nor affordability. for that, you really need to use an income figure.

    median income has dropped about 9% since 2009.

    add in the nominal price increases in thanksgiving dinner, and this looks significantly less affordable at the median than 3 years ago.

    price is up 15% from 2009 and income is down around 9%. that’s a pretty large change in affordability.

    also, does this thanksgiving basket look absurd to anyone else?

    14 oz of stuffing for 10 people? only a 16 pound turkey? 1 pound of peas? 1/2 pint of cream for 2 pies?

    outside of munchkin land, this does not seem like nearly enough food.

    i hosted thanksgiving last year for 7 and we spent roughly 2X this whole purported budget just on the turkey. (granted it was organic and heirloom)

    • 1. The post has been updated using average hourly earnings of all private workers instead of wages for production workers. Measured in time, the cost of the turkey dinner has decreased slightly, from 2.12 hours to 2.10 hours, and has therefore slightly “cheaper” this year in the “time cost.”

      2. We could also say that the average hourly wage has increased by 1.55% from October 2011 to October 2012, compared to the increased cost by 0.57%. Because average wages increased almost three times greater than the increase in the cost of the meal, we could say the meal is “cheaper” this year relative to income, or we could say it’s now slightly more affordable.

      2. If that’s not enough food, you can double or triple the basket of food, and the analysis wouldn’t change. The main point is to use a consistent basket from year to year and compare….

      • I agree that long term the hours needed to pay for an item is the only valid method. By that method of course as has been noted electronics are drastically cheaper than in the past. It would also be interesting to plot the hours to pay for a car over time. I recall the 75 anniversary edition of forbes did this for a number of items.

        • To take an example I have a letter from my grandfather that says a draftsman in 1920 made about $20/week and gasoline cost about .2/gallon, thus about 20 mins per gallon. Today according to SSA the average wage is about $21.5/hour. Thus at $4 its about 11 mins to buy a gallon, and less at $3 about 8.5 mins. So gasoline is cheaper than 1920. Also although I don’t have the numbers handy the hours to buy a kwh of electricity have come way down since 1920, I found an article at hemmings motor news: that says electricity cost about .0745/kwh in 1920, where as today its between .09 and .10 on average. Taking the numbers its about 9 mins in 1920 and about .27 mins today, which of course is a drastic drastic improvement. Note that the article says the average wage (based on 50 weeks of work) was about $30 in 1920. so the times would be 2/3 of the times I cite for 1920.

        • lyle-

          let’s say your boss cuts you back from 40 hours a week to 30 but increases your salary from $10/hr to $11 hr.

          that decreases the number of hours you need to work to buy a gallon of gas. but doe that really make it more affordable?


          your purchasing power dropped by 17.5%.

          hours worked to buy somehting is only part of the story. the number of hours that you do work matters too.

      • that hourly wage figure may not be a good metric either.

        if the workweek has shortened (and it has) a meal may be a larger portion of income even if it takes fewer hours.

        median income has dropped, not increased. the increase in hourly wages has been more than offset by a drop in hours worked/declines in salaries and bonuses.

        therefore, i would still argue that this meal less, not more affordable.

        you do not gauge affordability based on hours worked, but as a fraction of overall pay.

        by the logic you are using, i could get cut back to one hour a week at $100/hr vs my old job of 40 hours at $10 and that would make this meal more affordable.

        i do not see how that makes any sense.

        • Morganovich,

          If one is trying to show that government intrusion into the marketplace has reduced standars of living, then comparing food costs to monthly household income might be valid.

          However, if one is trying to show the miracle of markets in driving down food costs over half a century, then there is nothing wrong with comparing the cost of a Thanksgiving meal with the average hourly wage.

          IMO, Mark has provided an excellent metric for showing the latter. That he wasn’t intending to measure what you would measure doesn’t make his metric invalid.

          • My remarks were only aimed to the bottom chart. If 50% of food costs are spent away from home, that can’t be ignored in the overall picture.

            We do tend to forget the decrease of the time it takes to prepare food along with the decrease in time it takes to earn food.

          • Walt,

            My comment at this part of the thread, a response to Morganovich, was that the metric:

            “work hours required to pay for cost of Thanksgiving dinner”

            is valid for the period 1986 to 2012. That’s because the preparation of such a holiday dinner over that period has not likely changed.

            I agree that Mark’s bottom chart – showing spending at home as a percent of disposable income from 1948 to 2012 – is not really accurate. But not for the reason Morganovich gave about declining work week. It’s just that too many factors have changed over that 64 year period, including:

            - time required to prepare food in the home
            - percent of food consumed outside the home
            - demographic changes, such as age of the population

    • Your food budget or your supermarket budget? When we buy groceries at Walmart, we also buy shampoo, dog food, saran wrap, and golf tees. Do you separate and keep track of all the food items you buy?

      • That’s just for food. Other items are bought in bulk from dollar stores or Costco.

        We do help feed two teen age boys who are human vacuum cleaners. The per capita chart shows $11.52 per day per person for food from all sources in 2011. I think that is accurate from the people I talk to. Families are spending $200-300 per week around here at the grocery store (that would include non-food items). Average take-home pay on a joint income tax return is in the $30,000 to $50,000 range (I do income taxes). After buying food and gasoline many people are only paying the shut-off amount on their electric and gas bills every month.

  3. Concerning the Update: I can’t see how anyone with a common income can only spend 5.7% of their income on food ($10 per day per person is more like it or about $140 per week for 2 people). The numbers just don’t work out to 5.7% for anyone with an income of $100,000 or less. I think common sense is lost in the aggregate in table 7.

    Income % food Year Food Week Food
    $100,000.00 0.057 $5,700 $109.62
    $80,000.00 0.057 $4,560 $87.69
    $60,000.00 0.057 $3,420 $65.77
    $40,000.00 0.057 $2,280 $43.85
    $20,000.00 0.057 $1,140 $21.92

    • The data are not reporting spending on a food as a share of household income, they are reporting total food spending as a share of total personal disposable income – at the aggregate level, not the household level. Also, that’s only on food at home, there’s another category for food away from home, i.e. restaurants.

      • Thanks for the explanation, Mark. Sometimes when we see numbers like yours in a chart we question their usefulness because we cannot see how they realistically apply to us. I guess data in that form are useful for people who make policy decisions at a high level. I can see how the downward trend over time would be useful.

        No one I know only spends 5% of any type of their income on food purchased at the grocery store–it’s two to three time that much for most of us. Wouldn’t a percentage of food spending as a share of household income be a better metric for a general reading audience? Just asking :)

        • From Table 13 USDA data on food expenditures, Year 2011:

          $41.75 At home food per person per week
          $39.58 Away from home food per person per week
          $81.33 Total food per person per week

          I was not aware people spend about 50% of their food budget away from home. I am a cheap ass.

          • Many of us spend way more on meals away from home than on meals prepared at home. One meal at Ruth’s Chris costs as much as all the grape nuts, bagels, and pop tarts I consume in half a year.

          • I was surprised at the increase of the share of income spent for food away from home. When that is figured in, the share of money spent on food has decreased from about 22% to about 11% since 1949 (or halved). That’s still something to be thankful for but not as quite of a dramatic drop as shown in the chart above.

          • Mark’s comparison of hours worked to pay for a Thanksgiving dinner makes sense to me.

            Any comparison of grocery spending now vs 50 or 60 years ago is not valid, IMO. As housewives moved into the paid work force, the preparation of food was moved from highly talented artisans to mass production workers. The prices we pay for most grocery items today represents not just food costs but also the preparation costs.

            Unless we account for the many hours which housewives spent preparing food 60 years ago, we have understated the cost of 1949 food when we try to compare it with 2012 food.

  4. John

    As housewives moved into the paid work force, the preparation of food was moved from highly talented artisans to mass production workers.

    You obviously never ate at my house as a kid.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>