TradeAlmanac
Sign in

The consumer basket: whose shopping list Rosstat measures and why it isn't yours

· 5 min · beginner

Automated material · TradeAlmanac editorial deskDraft prepared by a language model from our stored data; not reviewed by an editor.

66 views
The consumer basket: whose shopping list Rosstat measures and why it isn't yours — Macroeconomics

The consumer basket is not a shopping list for getting by — it is a set of weights. Rosstat fixes a list of goods and services and the share each item takes up in the spending of an average household, then tracks how prices change for exactly that set. The result is the consumer price index: the inflation figure the Bank of Russia watches, the one used to index payments, and the one to which the face value of certain government bonds is linked. The question "what's in the basket" nearly always means one of two different questions: about the statistical weights of the CPI, or about the social standard that used to underpin the subsistence minimum. These are different constructs with different fates.

Two different baskets under one name

The first basket is statistical. Its job is to measure price changes, so it tries to reflect actual consumption rather than what is desirable or sufficient. If people spend a lot on mobile services, mobile services get a noticeable weight, regardless of whether anyone considers them a necessity.

The second basket is normative. Until recently, Russia's subsistence minimum was calculated from it: the law set a minimum set of food products, and non-food goods and services were added on top in a fixed proportion. That is where the persistent image of the "basket" as a list you can live on for a month comes from. This mechanism has been abolished, and there is no longer any point in confusing it with the statistical set.

How the statistical basket works

The set is split into broad blocks: food products, non-food goods and services. Within each block sit specific representative items: not "bread in general" but a particular type of loaf of a particular weight; not "travel" but a trip on a specific mode of transport. Price collectors record prices for these representative items at retail outlets across the regions, and the average price of each item is built up from a large number of observations.

The weights come from the household budget survey — a poll of what families have actually spent their money on. Weights are revised when the structure of spending shifts; the list of items is also updated, with disappearing products dropped and new ones added. The index formula itself follows the Laspeyres approach: prices change, while the structure of consumption within the period is held constant. Hence a well-known property: such an index ignores the fact that when prices rise, people switch to cheaper substitutes, and so it tends to slightly overstate the growth in the cost of living. To separate underlying price movement from one-off spikes, Rosstat also publishes core inflation, which strips out the most volatile components — above all fruit and vegetables, fuel and regulated tariffs.

It helps to keep the terminology on this topic close at hand — definitions of the CPI, core inflation and indexation are collected in the site glossary.

Why personal inflation always diverges from the index

Rosstat's basket describes an average household that does not exist. Your personal basket has a different structure: the share of food falls as income rises, a tenant and a homeowner live under different price regimes, a driver is sensitive to fuel, and a family with a child to clothing and educational services.

So the feeling that "prices have risen more than the statistics say" usually points not to a calculation error but to a mismatch in weights. The psychology of perception amplifies the effect: price rises on frequent small purchases stick in the memory, while stable or falling prices for electronics and telecoms do not.

What changed in 2021

The subsistence minimum is no longer calculated from a set of products. Amendments adopted at the end of 2020 switched the method to a relative principle: the minimum is defined as a share of median per capita income for the previous year, and the minimum wage is tied to the median salary. The law on the consumer basket for the Russian Federation as a whole ceased to apply.

The point of the switch lies in a change of logic. The normative set answered the question "how much is needed to survive physically"; the relative criterion answers the question "how far has a person fallen behind the typical standard of living in the country". Meanwhile, the statistical CPI basket has not gone anywhere: it continues to be calculated and remains the basis for indexation.

Where investors encounter the basket

Three mechanisms make this topic a practical matter rather than mere background knowledge.

Step one: the rate. The Bank of Russia targets inflation as measured by this basket. The deviation of actual inflation from the publicly announced target is the main argument in key rate decisions, and the key rate reprices the entire yield curve in the debt market.

Step two: inflation-linked debt. The face value of OFZ-IN bonds is linked to the consumer price index with a lag of several months, which means it protects against precisely the inflation Rosstat has recorded — not against the rise in your personal basket. Issues and their terms are easy to browse in the OFZ section.

Step three: corporate revenue. For retailers and food producers, the average ticket grows along with basket prices, but costs and regulated tariffs move at their own pace — hence the gap between nominal revenue and margins, which shows up only in financial statements, not in the index. You can assess this for specific issuers in the equity market.

Where to find current figures

This text deliberately contains no current values: any inflation figure goes stale by the next release. Release dates for statistics and Bank of Russia meetings are in the events calendar, and the latest numbers are in the news feed. The composition and weights of the basket for a given year are published by Rosstat itself; the site does not carry that breakdown, and passing off approximate shares as exact ones would be worse than honestly referring readers to the primary source.

Share
Was this useful?
How this material was prepared

Draft prepared by a language model from our stored data; not reviewed by an editor.

Model: claude-opus-5

How we use language models

Similar articles