MisleadingCharts
Back to the gallery

The record cheese mountain that would last 40 days

Showing the misleading chart

Be the first to star this exhibit

A mouse emigration notice we drew from USDA cold-storage figures celebrates America’s tallest cheese pile ever: 1.52 billion pounds in July 2022, 14 all-time highs since 2016, the 1983 “government cheese” summit left behind. Divide the pile by how fast the country makes cheese and the record is 40 days of output, against 96 days in 1983 and 161 in 1954.

01The claim

America’s cheese mountain has never been taller. A notice of the kind a field-mouse emigration society might pin up, which we drew from USDA survey figures: every pound of natural cheese in U.S. cold storage at the end of every published month since 1920, one zero-based axis, nothing smoothed. The pile reached 1.52 billion pounds in July 2022. It first beat the 1983 “government cheese” summit of 1.25 billion in May 2016, the first of 14 all-time highs since then. Recommendation: book the autumn crossing and pack the large bag.

02The trick

Every figure on the notice is right, and the axis is beyond reproach. What is missing is the thing a stockpile has to be measured against: how fast it gets used. A pile of cheese is only large or small next to the rate the country makes and eats it, and that rate has not stood still. Over the twelve months to September 1983, U.S. plants made 4.77 billion pounds of cheese; over the twelve months to July 2022 they made 14.04 billion, 2.9 times as much. The pile grew 22% over the same span. So the 1983 summit held 96 days of the country’s output and the 2022 “record” holds 40, and by that measure July 2022 sits below every summer from 1950 to 1987. Raw totals of anything measured against a growing base will keep setting records on their own, the way a big country tops every list. The two piles were also different things. By 1984 about 60% of cheese stocks were owned by the federal government, bought to support milk prices. The 2022 pile was commercial inventory: under six weeks of output for a much larger industry. Same word, same axis, different mountain.

03The fix

Divide before you compare. Put the stockpile over the flow it serves, here end-of-month stocks over the past twelve months of production, expressed as days on hand, and draw that on a zero-based axis. The honest chart shows the numerator and the denominator as small panels beside the ratio, so a reader can see where the change comes from: stocks up 22%, output up 2.9 times. The ratio tells a different story from the pounds. (Production here is NASS’s total cheese, which is natural cheese excluding cottage, so numerator and denominator match in kind.) Cheese stocks stood at 161 days of output in 1954, 96 in 1983 and 21 in November 1995, and since 2000 they have stayed between 28 and 41 days. The July 2022 “record” is 39.6 days, and August 2026 is 35. Divide by domestic use instead of production and July 2022 is about 41 days of 2022’s use (USDA ERS supply and utilization tables), so the answer does not hinge on that choice. It is days of output, not days of supply: 7–9% of U.S. cheese output in 2022–25 was exported. Then say what the pile is. USDA notes that the federal government owned about 60% of cheese stocks in 1984, while today’s stocks are almost entirely privately held, and some of any cheese inventory is cheddar left to age on purpose. Both 1983 and 2022 show a lot of cheese, but only the 1983 pile was largely bought up by the government.