The simple version

The figure in question answers a narrow question: in the places that were already trading a year ago, is more money coming in than came in then.

It only answers that question usefully if money means roughly the same thing in both years. Where prices have run away, it does not.

So the definition quietly drops those countries. The number that reaches the headline is built from the countries left, and the headline does not mention the ones taken out.

The numbers

• The definitions page of McDonald's Corporation's third quarter 2026 results release states that comparable sales exclude the impact of currency translation and the sales of any market considered hyperinflationary, generally identified as those markets whose cumulative inflation rate over a three-year period exceeds 100%, which management believes more accurately reflects the underlying business trends.
• The same page defines comparable sales as sales at all restaurants, whether owned and operated by the company or by franchisees, in operation at least thirteen months, including those temporarily closed.
• For the quarter ended September 30, 2026 the company reported global comparable sales up 1.5 percent, with the United States down 0.2 percent, International Operated Markets up 1.4 percent and International Developmental Licensed Markets up 4.2 percent.
• Neither the release nor the Form 10-K for the fiscal year ended December 31, 2025 names any market treated as hyperinflationary in this quarter. We looked in both and found no list, so nothing here says that any particular country was left out.
• The figures come from exhibit 99.1 to a Form 8-K filed November 4, 2026, accession 0000063908-26-000080.

Why a year-on-year reading stops working when prices run

A year-on-year reading is a ratio between two money amounts measured twelve months apart. It is only a statement about trade if the unit of measurement held still.

Where prices have tripled, the unit did not hold still. The same amount of trade across the counter produces a far larger money figure, and the ratio reports an enormous rise that nobody lived through.

Worse, that one country can swamp the rest. A reading built from many countries is an average, and a country posting a figure many times larger pulls the average up far enough to bury what happened elsewhere.

Taking those countries out is therefore the more informative choice rather than a flattering one, and the company says as much. The cost of making it is that the set of countries behind the figure can change without the figure announcing it.

The actual math on a country where prices triple

Picture one place, in a country where prices triple over the period, putting exactly the same number of plates across the counter as it did a year earlier. The plate count, the tripling and the amounts below are ours, chosen to show the mechanism.

Say it took in 1,000,000 units of that country's own money a year earlier. With prices tripled and the same plates across the counter, it takes in 3,000,000 now.

That reads as a rise of 200 percent, from a kitchen that handled not one extra customer. Average that against a country growing 2 percent and the pair comes out near 101 percent, which describes neither of them.

What this means

A headline growth figure is only as meaningful as the set it was built from, and that set is defined in small print that can move. Reading the definition is the only way to know what is in and what is out.

The broader point is that any average across countries carries the currency problem inside it. Where prices move at very different speeds, an unadjusted average mostly reports the fastest mover.

The transferable habit is to ask what a reported figure leaves out before comparing it with last year's version of itself. A rule that removes members from a group will eventually remove one, and the figure looks the same either way.

What this is NOT

This article explains why a company's definition of comparable sales leaves out markets it considers hyperinflationary and what that does to the reported figure, using the definitions page of McDonald's Corporation's third quarter 2026 results release, exhibit 99.1 to a Form 8-K filed November 4, 2026, read on November 4, 2026. It does not say that any market was treated as hyperinflationary in this quarter or in any other: neither document we read names one, and we found no list. The single place, the tripling of prices, the 1,000,000 units of money and the 2 percent country in the math section are all ours and appear in no filing. It is not a recommendation to buy, sell or hold any security, not a view on any company, and not a suggestion that the practice is misleading: the rule is stated in the release and the company gives its reasoning. It makes no forecast of inflation, prices or sales in any country, and it is not economic advice about any country's situation. This is not a political endorsement or criticism of anyone.

Sources

• McDonald's Corporation, exhibit 99.1 to Form 8-K, third quarter 2026 results, filed with the Securities and Exchange Commission November 4, 2026, accession 0000063908-26-000080: the definitions of comparable sales and systemwide sales, the hyperinflationary market language and the three-year cumulative inflation threshold, and the reported comparable sales figures by segment. Read November 4, 2026: https://www.sec.gov/Archives/edgar/data/0000063908/000006390826000080/exhibit991-9302026.htm
• McDonald's Corporation, Form 10-K for the fiscal year ended December 31, 2025, accession 0000063908-26-000035: checked for a list of markets treated as hyperinflationary, and none found. Read November 4, 2026: https://www.sec.gov/Archives/edgar/data/63908/000006390826000035/mcd-20251231.htm