Brief Description of Retail Food Price Index

The Retail Food Price Index, a component of the Consumers' Price Index, measures average changes in retail prices of a fixed list of foods of constant quantity and quality, bought by moderate-income families in large cities. This is in line with the general purpose of the Consumers' Price Index of measuring how much more or less it costs at one time than at another to purchase a fixed list of goods. The index is not designed to measure how much more it costs to live in one city than in another.2

Retail food prices were first collected in 1903, when the Bureau's representatives obtained prices for the years 1890 through 1903 from grocers' records. At that time, 30 foods were priced in 171 representative cities in 33 States. Since then changes in the lists of foods and in the number of cities have been made, with the number of foods varying between 16 and 87 and the number of cities between 39 and 171. The base period, collection and computation methods, and techniques have also changed from time to time.

Currently the Bureau publishes retail prices of 50 foods in 56 cities. Each month about 80,000 quotations are collected from 1,650 independent stores and 150 chain organizations representing 6,500 chain stores, or a total of about 8,150 stores.

Store Sample Selection

In selecting the sample of stores for food price reports, the Bureau has taken into account type of store in terms of foods handled, size of store as measured by sales volume, and geographic location within the city.

Revisions in store samples are made from time to time, to maintain the accuracy of the Bureau's food price index. The latest complete sample revision took place between September 1945 and

1 A detailed discussion of the Consumers' Price Index will be presented in a forthcoming bulletin, Consumers' Prices in the United States, 1942-48 (Bull. 966). The index as it was computed through 1941 is described in Changes in Cost of Living in Large Cities in the United States 1913-41 (Bull. 699).

' A special study of differences in costs between cities is presented in The City Worker's Family Budget in the Monthly Labor Review, February 1948 (also reprinted as Serial No. R. 1909).

June 1946. At that time the size of the sample of independent stores in each city was changed so as to be equal to the square root of the total number of independent food stores operating in the city. This relationship was employed since the ratio necessary to obtain stable average prices in a small city is higher than is necessary for a large metropolitan area.

A complete listing of all independent stores in each of the 56 cities was classified according to type of commodities handled-combination stores (groceries and meats), groceries only, meats only, produce markets, etc. The listing for each store type was further classified by sales volume classunder $50,000, $50,000 and under $250,000, and $250,000 and over annual sales volume. Stores were further distributed within city areas and a random selection then made within each area to fulfill the sample requirements. The result was a self-weighting sample of independent stores based on current distribution of total independent store sales in each city. The Bureau continued to include all important chain stores in each city. Collection of Prices

The Bureau collects retail prices of 50 foods in each of the 56 large cities included in its Retail Food Price Index, during the first 3 days of the week containing the fifteenth of the month. Local Bureau representatives collect retail food prices from grocers who report voluntarily. The representatives are provided with a description (specification) of the quality for which price quotations are desired. Within the range of each specification, they are instructed to secure a price for the type, brand, etc. that is sold in greatest volume in each store. Specifications are defined precisely enough to insure a meaningful average price and avoid movement in the index because of shifts in the quality priced from one period to the next. They are also broad enough, within limitations, to provide an adequate number of quotations and to allow for city and regional differences in grades, types, package sizes, etc.

* See Store Samples for Retail Food Prices in Monthly Labor Review for January 1947; also reprinted as Serial No. R. 1878.