The time price of food is what your groceries cost in minutes of your life rather than in dollars: divide the price by your hourly wage, and read the answer in minutes. Run it on the cheapest ground beef in America and you get about 9 minutes 15 seconds a pound in 1984, 6 minutes 15 seconds in 1998, and 12 minutes 45 seconds today. The dollar figures on their own never told you that, because the dollar is not a fixed ruler.
I have had some version of this argument a dozen times, usually standing in a kitchen holding a receipt.
I say food has gotten more expensive. Someone tells me I am being nostalgic — that Americans spend a smaller share of their income on food than they ever have, that this is one of the great achievements of the modern economy, and that my grass-fed brisket habit is a lifestyle choice rather than an economic observation.
They are not wrong about the statistic. They are wrong about what the statistic measures.
Why food prices in dollars cannot be compared across decades
Here is the problem in one sentence: when you price food in dollars, you are measuring a thing that changes with a ruler that also changes, and you have no way to tell which one moved. It is the oldest trick in monetary economics, and it is not even a trick — nobody is hiding it. We just stopped noticing.
A dollar is not a unit like a meter or a gram. Nobody keeps a reference dollar in a vault in Paris. The dollar is a claim on an expanding pool of credit, and its purchasing power is an output of policy, not a constant of nature.
This is not a fringe view. It is the boring, textbook description of a fiat monetary system. The only thing sound-money people do differently is take it seriously enough to stop using the dollar as a measuring stick.
Because once you accept that the ruler moves, every price comparison across time becomes suspect. “Beef cost $1.29 a pound in 1984 and $6.89 today” tells you almost nothing on its own. It tells you the two numbers. It does not tell you whether beef got scarcer, whether the money got weaker, or — most likely — some combination that the single number can never decompose.
I wrote about the mechanism behind that in How Money Broke Food. This piece is about the measurement. Specifically, it is about a fix you can run yourself, on your own receipts, this afternoon.
What is a time price, and how do you calculate one?
Divide the price by what you earn in an hour. Report the answer in minutes. That is the whole method.
It has a name, and the name is not mine. Marian Tupy and Gale Pooley call it a time price, and they built a book around it — Superabundance, published in 2022. Their formulation is exactly the arithmetic above: Americans buy goods with money but pay for them with time, so divide the dollar price by the hourly wage and read the result in minutes. I will come back to them further down, because they run this same calculation on food and reach a conclusion opposite to mine.
It sounds almost too simple to be useful, and for a while I assumed someone would tell me why it was naive. Nobody has. What it does is swap out a denominator that is being actively managed for one that is anchored to something real: the time you had to spend to get the thing. Time is the one input that cannot be printed, which is exactly why it makes a decent ruler.
Austrian economists have a name for the underlying idea — the pure time preference at the heart of all exchange, which I got into in Time Preference and the Plate. But you do not need the theory to run the arithmetic.
The time price of ground beef, 1984 to 2026
Let me use the most ordinary food in America, and the most boring possible wage series, so nobody can accuse me of picking flattering inputs.
The price: BLS average price for 100% ground beef, per pound, U.S. city average (series APU0000703112).
The wage: BLS average hourly earnings for production and nonsupervisory employees, total private (series AHETPI). This is roughly the non-management four-fifths of private-sector workers — not the average including executives, which flatters the result.
Three snapshots, each pairing the same month:
- January 1984 — beef $1.29, wage $8.38/hr. About 9 minutes 15 seconds of work per pound.
- September 1998 — beef $1.36, wage $13.11/hr. About 6 minutes 15 seconds.
- July 2026 — beef $6.89, wage $32.42/hr. About 12 minutes 45 seconds.
Late 1998 is, as far as I can find in this series, the cheapest ground beef has ever been in American labor-hours. Since then the cost in time has roughly doubled. Against 1984 it is up by about 38 percent.
Notice what the dollar figures alone would have told you: beef is up 5.3x since 1984. Notice what the wage figures alone would have told you: pay is up 3.9x. Neither number is the answer. The ratio is the answer, and the ratio says a working person now trades meaningfully more of their life for the same pound of the cheapest beef on the shelf than they did when I was a kid.
Two honest caveats about the data
I would want these if someone handed me this. First, BLS says plainly that its average-price series are best used to measure the price level in a given month rather than change over time — the item definition and outlet mix drift, and the CPI index for beef is the more rigorous instrument for change. I use average prices here because they are what a shopper actually confronts, and I would rather be transparent about the tool than pretend to a precision I do not have. Second, AHETPI is a nominal wage series with a known artifact: it jumped in April 2020 not because anyone got a raise but because low-wage workers were laid off first, which mechanically lifted the average. Neither issue moves the doubling-since-1998 result, but you should know they are there.
But is food not cheaper than it used to be?
The strongest counterargument is the one I opened with, and it deserves a real answer rather than a dismissal. It arrives in two forms.
The first is the spending share. USDA’s Economic Research Service puts food at 9.7 percent of disposable personal income in 2025, down from 10.4 percent in 1997 (ERS Food Expenditure Series). By that measure food is cheaper. That is a real number from a serious agency and I am not going to wave it off.
But look at how it decomposes. The entire decline came from food at home, which fell from 6.1 percent to 4.8 percent. Food away from home actually rose, from 4.3 to 4.9 percent. Americans did not get cheaper groceries and bank the difference. They shifted spending toward restaurants and prepared food while the grocery line got thinner.
The second is a time price that points the other way — and this is the one I have to take most seriously, because it uses my own method against me. Tupy and Pooley apply time prices to the American Farm Bureau Federation’s annual survey of a Thanksgiving dinner for ten, and find that the meal cost 3.22 hours of blue-collar work in 1986 and 1.76 hours in 2025, a decline of 45.3 percent (Human Progress). Same arithmetic, same family of wage series, opposite direction.
Both results are real, and what separates them is the basket. Theirs is a mixed haul of a dozen or so items, most of them precisely the commodity calories that fiat-era agriculture is best at producing cheaply. Mine is a single item, and beef has a supply story of its own at the moment: the U.S. cattle herd stood at 86.2 million head on January 1, 2026, the smallest in 75 years (Drovers). Some of that 12 minutes 45 seconds is monetary and some of it is cattle, and I cannot cleanly separate the two.
That is not a retreat, it is the point. A time price gives you an honest denominator. It does not tell you why the numerator moved, and it cuts both ways — if the method says the Thanksgiving basket got cheaper in hours, then it got cheaper in hours, and I should say so rather than reach for a different measure.
What the aggregate cannot do is answer the question I am actually asking, which is about the specific thing I am trying to buy: pastured meat from a named farm ninety minutes away. Cheap calories got dramatically cheaper. That is genuinely an achievement, and it is also why the average can fall while the item on my list gets harder to afford in hours. Both things are true. The aggregate is not lying; it is answering a different question. I made the parallel case on the nutrition side in Are Vegetables Less Nutritious Than They Used to Be?
Where the money in your food dollar actually goes
One more number worth carrying around. Of every dollar Americans spent on domestically produced food in 2024, farms received 11.8 cents (USDA ERS Food Dollar Series). On food at home it is 18.5 cents; on food away from home, 7.1 cents. The other 88 cents is the marketing bill — processing, packaging, freight, wholesale, retail, advertising.
I find this clarifying rather than depressing. When you buy a quarter cow direct, or hand cash to a grower at a market stand, you are not just paying a higher sticker price for better food. You are collapsing a supply chain that was consuming most of the dollar. Whether that arithmetic works out in your favor depends on the specific deal — I ran the actual SoCal numbers in What a Quarter Cow Actually Costs — but you should at least be comparing the right things.
How to price your own groceries in hours
Take your own hourly number. Not your salary — divide it out, and be honest about the hours you actually work.
Then price things in minutes. A dozen pastured eggs. A quarter cow, amortized per pound. A CSA share divided by the weeks. The rotisserie chicken. Compare them to each other in minutes rather than dollars, and compare them to what you remember paying.
Two things tend to happen. The gap between industrial and real food narrows more than people expect, because the real-food premium is a ratio and ratios do not care what currency you quote them in. And the drift in the dollar itself becomes visible, because you finally have a denominator that does not move when the Fed does.
That second effect is the one that changed how I think. It is the same reason I keep a portion of savings in a fixed-supply asset and write about the monetary side of this at Modern Wealth Model and Halving House. Money that holds its length lets prices carry information again. Prices that carry information let you tell the difference between food that got expensive and money that got weak.
Until we have that, do it by hand. Divide by your wage. Count in minutes. It costs nothing and it is very hard to argue with.
Frequently asked questions
What is the time price of food?
The time price of food is its cost expressed in working time instead of currency: the dollar price divided by your hourly wage, reported in minutes or hours. Because the wage and the price are both quoted in the same dollars, the dollar cancels out — which is the point, since the purchasing power of a dollar changes over time and the length of an hour does not.
How do you calculate a time price?
Take the price of the item, divide it by your after-tax hourly earnings, and multiply by 60 to get minutes. A $6.89 pound of ground beef against a $32.42 hourly wage is 0.2125 hours, or about 12 minutes 45 seconds of work.
How many minutes of work does a pound of ground beef cost?
Using BLS average prices against the production and nonsupervisory wage: roughly 9 minutes 15 seconds in January 1984, 6 minutes 15 seconds in September 1998, and 12 minutes 45 seconds in July 2026. The 1998 figure is the lowest in the series as far as I can find, which means the time price has roughly doubled since then.
Is food more expensive than it used to be?
It depends entirely on which food. Measured as a share of disposable income, or as a mixed grocery basket priced in hours, food has become more affordable. Measured one item at a time, the answer varies enormously — and for beef specifically the time price has roughly doubled since 1998, driven by both monetary debasement and a cattle herd at a 75-year low. The aggregate cannot answer the question for the item on your list. Only the arithmetic can.
How much of my food dollar goes to the farmer?
About 11.8 cents of every dollar spent on domestically produced food in 2024, according to USDA ERS — 18.5 cents for food eaten at home, 7.1 cents for food away from home. The remaining share is the marketing bill: processing, packaging, freight, wholesale, retail and advertising.
The same caution applies one level down, to what the package claims about the animal rather than what it costs. USDA stopped defining “grass-fed” in 2016, and certified organic beef can legally be grain-finished in a feedlot — I sorted the whole labeling system in What Meat Labels Actually Mean.
Find growers you can buy from directly in the Producers directory, or a stand near you in Markets.
Related: land is the other side of the same question. See Is Farmland a Hard Asset? What 55 Years of USDA Data Say.