The Index Sees It, the Shopper Does Not: Shrinkflation, Sticker Shock and the Limits of Attention

Two complaints recur in Canadian conversations about the cost of groceries. The first is that packages have become smaller while the price on the shelf holds steady. The second is that shelf prices, when they do move, move further than expected. Both are commonly folded into a single grievance: that the official inflation figure understates what households actually face.

The two complaints describe different phenomena, and they stand in different relationships to the measure. One of them is captured by the Consumer Price Index in full, and more completely than most shoppers capture it themselves. The other is recorded accurately and then diluted. What follows is an account of both mechanisms.

How the index handles a smaller package

Statistics Canada treats a reduction in package size as a price increase. The agency states that when the quantity or size of a product is reduced while the price stays the same, consumers are paying more for the same quantity of the product. To account for this, the prices collected are adjusted upward to reflect the change in quantity, and the resulting price increase is reflected in the index as a pure price change. The agency refers to the process as quantity adjustment.

The principle underneath it is older than the term. The Consumer Price Index measures price change by comparing the cost of a fixed basket of goods and services through time, and because that basket holds quantity and quality constant, the index reflects only pure price change. Where prices relate to different quantities, they are adjusted to a standard quantity unit before any comparison is made.

The arithmetic is not complicated. A product reduced from 200 grams to 180 grams at an unchanged shelf price has become roughly 11 percent more expensive per gram, and an increase of that order is what the index records.

The scale of it, measured

In February 2025, Statistics Canada published a dedicated analysis of food-specific quantity adjustments. Between 2021 and 2023, 29.6 percent of eligible grocery items tracked in the Consumer Price Index experienced shrinkflation.

The qualifier carries weight. Eligible items are representative products, and the category excludes all fresh fruits and vegetables and most meat products, so the figure describes packaged goods rather than the whole basket. The period matters too: the analysis runs to 2023 and is not a current reading.

Within that window, the concentration is pronounced. The distribution ran 24.4 percent in 2021, 49.5 percent in 2022 and 26.2 percent in 2023, shares that do not sum to 100 percent owing to rounding. Nearly half fell in 2022, the year grocery inflation reached its highest annual average in 41 years at 9.8 percent, a rate applied to what a household spends on groceries over a year.

Name-brand products accounted for 77.6 percent of instances against 22.4 percent for house-brand products. The foods most often affected were margarine, pasta mixes, cookies and crackers, mozzarella cheese, breakfast cereal and other cereal products, and cheddar cheese.

Why the increase does not register

Shoppers hold a reference price in memory for familiar items and compare against it. The comparison is fast and serves its purpose well, but it is a comparison of prices rather than of rates. When quantity falls while price holds, the remembered reference survives intact and the comparison returns no change, even though the cost per unit has risen by the full amount the index has just recorded.

The information required to detect the change is present at the shelf. Net quantity appears on the package and unit price appears on most shelf labels. Neither is the figure the comparison is being made against. The Office of Consumer Affairs at Innovation, Science and Economic Development Canada treats per-unit pricing as the practical response, and addresses shrinkflation and skimpflation alongside it.

The brand split reported by Statistics Canada gives the point additional force. Where shrinkflation is concentrated in name-brand products, a shopper comparing unit prices rather than package prices is comparing on precisely the dimension along which the two ranges have diverged.

The opposite failure

Sticker shock runs the other way. A price that jumps between visits registers immediately and is remembered well. The index records the movement accurately, and then sets it among everything else it tracks. The Consumer Price Index draws on some 491 categories of goods and services, used to represent price movement in 220 lowest-level classes. A shopper examines a handful of them.

Frequency drives salience. Statistics Canada has observed that consumers are more likely to notice, and to attach greater importance to, price changes for the things they buy frequently than for occasional purchases. Groceries are bought weekly, and their prices are re-encountered constantly, so grocery price movements are recalled with unusual clarity; the weekly grocery budget households plan around is among the few costs watched that closely. A category bought once a year, or paid by automatic transfer, leaves no comparable impression regardless of how far it moves.

The resulting mental index is weighted by attention rather than by expenditure. It is not mistaken about the items it contains. It is assembled from a far smaller sample.

What remains unsettled

The statistical treatment of shrinkflation is settled. Whether shoppers should be told about it at the shelf is not.

In June 2026, the Competition Bureau announced an examination of competition across Canada's food supply chain, covering production, distribution and retail pricing practices, including loyalty programs, pricing algorithms, shrinkflation and skimpflation. The Bureau has stated that this is neither a market study nor a law enforcement investigation, and no new legal requirement has been proposed. Its published questions ask how shrinkflation and skimpflation affect the ability of consumers to compare prices and value across retailers, and what measures might improve transparency in retail pricing. The submission period was extended to 4 September 2026, with a report expected in spring 2027.

Existing law addresses quantity but not change. The Consumer Packaging and Labelling Act requires prepackaged products to carry a declaration of net quantity, prominently displayed and legible. Nothing requires an indication that the quantity has changed, and a smaller package at a constant price reaches a household grocery budget whether or not it is noticed, which is one measure of how persistently grocery costs have outpaced the broader index.

A problem of attention

The inversion is worth stating plainly. On shrinkflation, the index is the more sensitive of the two. It converts a smaller package into a price increase and records it in full, while the shelf price holds steady and the shopper's comparison returns nothing. On sticker shock, the shopper is the more sensitive. A single vivid movement dominates recollection, while the index files it among hundreds of categories and moves on.

The gap between the official figure and lived experience is real. On these two questions, however, it is not produced by anything the statistical agency does or fails to do. It is produced by what attention registers, and by what it does not.