Do inflation numbers account for shrinkflation?

And can shrinkflation be accurately measured?

By “shrinkflation”, you mean keeping the price per package the same, but making the package smaller? There are a lot of ways to measure inflation, but all of them are going to be based on actual measurements, not just “one package” of anything.

Yes, the official price counters at the Bureau of Labor Statistics account for product size:

Getting less for the same price? Explore how the CPI measures “shrinkflation” and its impact on inflation : Beyond the Numbers: U.S. Bureau of Labor Statistics (

BLS strives to capture product upsizing and downsizing in the CPI in a timely manner. Product downsizing and upsizing affect the cost to consumers of goods and services. The effective price changes due to size changes are reflected in the CPI. This analysis indicates that while only a small number of prices in the CPI experience downsizing each year, it does affect the indexes for specific item categories where such size reductions are common. However, the impact of product downsizing at the all commodity and services level is minimal, with an average annual effect of 0.01 percent per year, so while consumers may notice shrinkflation at the grocery store, it has a very small impact the overall inflation picture they face.

Interesting article. I had no idea that the BLS had economists whose job it was to track package sizes.

It’s the price per gram/ounce that matters.

This was my thought too. When CPI tracks “cost of soft drink” or “cost of bread” they presumably are measuring by the ounce or litre, not a specific product or packaging. Also, averaged over a large number of sources, so that one particular manufacturer and retailer does not horribly skew the results.

I wonder too if there are industrial statistics at play - the overall production vs revenue of say, the entire bakery industry or the soft drink industry.

Then there’s serving size over time, another type of inflation… which leads also to human inflation.

But package size has to be considered to some degree. A 1 ounce bag of Cheetos will cost more per ounce than a 1 pound bag.

And there’s the opposite strategy of serving size deflation in order to show a decreased number of calories per serving.

The inflation shopping basket takes averages. They don’t list “Cheetos” but “Breakfast cereals”.

The quantities or “weight” of the various items in the basket are chosen to reflect their importance in the typical household budget. There are, for example, several items in the basket covering purchases of bread – such as a large white sliced loaf and large wholemeal loaf – that are combined together to estimate the overall change in bread prices.

It is not, I hope, a breakfast cereal unless you’re an Oompa-loompa or someone else with a distinctly orange skin…?

I recall a consumer show complaining about this decades ago - when the labels indicating nutrition were first mandated, some companies were fudging their “suggested servig size”. For example a single-serving size of potato chips or candy, where the bag indicated the calories in the “suggested serving” was for 2/3 the contents of the bag. If you didn’t read carefully, you’d think it was 200 calories in the bag, not 300.

PAM Cooking Spray (and probably all cooking sprays) are famously 0 calories, despite being 100% fat, because the serving size is “1/4 second spray”.

They also do a great deal of work accounting for change in the quality of products. For example, a $1500 laptop purchased today will be much better than a $1500 laptop purchased a decade ago. The price index should reflect the fact that laptops are effectively falling in price over time.

This blog post explains the process in some detail:

Hedonic Models in the Producer Price Index : U.S. Bureau of Labor Statistics (

They now do this for a wide range of products, including some you wouldn’t expect:

Quality Adjustment in the CPI: Men’s Underwear : U.S. Bureau of Labor Statistics (

I assume that they are specific to N America.

I think that most inflation indices also weight necessities more heavily than luxuries. Everyone’s got to eat, but we don’t have to eat Cheetos. So even if a person eats more Cheetos than rice, the price of rice is still a better inflation measure (if you can’t afford Cheetos, you can still afford rice, and if you can’t afford rice, you can’t afford anything).

The indices are an attempt to measure price increases on the goods a typical (average?) person will buy. They will include basics like potatoes and rice, but also air fryers and new cars.

Obviously, since this is an average, it means that the index relates to middle-income people, rather than the poor or the rich.

The most widely quoted US CPI numbers do not in fact weigh things differently because they are necessities or luxuries.

In fact the second most commonly quoted CPI definition EXCLUDES Food & Fuel which most people would regard as necessities.

The most commonly used EU inflation measure excludes Owner Occupied Housing costs.