Volume 1: Consumer Price Indices

India (All India 2)

Official title

Consumer Price Index for Urban Non-Manual Employees.

Scope

The index is computed monthly and covers families deriving the major part of their incomes from non-manual occupations in the non-agricultural sectors, living in 59 selected urban centres.

Official base

April 1984-March 1985 = 100.

Source of weights

The weights and selected items were derived from a household expenditure survey conducted during one year from July 1982 to June 1983 among a sample of 45,000 non-manual households living in 59 selected urban centres. The weights were not adjusted to take into account price changes between the survey and base periods. The weights of items excluded were imputed to items included in the weighting diagram, generally on the principle of similarity in their price trends. If imputation to any specific items was not considered suitable, imputation was made at the sub-group level.

Weights and composition

Major groups Number of items (a) Weights (b) Approximate number of price quotations
Food and tobacco .........
Fuel and light .........
Housing .........
Clothing, footwear and bedding .........
Miscellaneous .........
Total ......167540

Note: (a) The number of items varies from 146 to 345 according to the centre.

(b) The weights differ in each centre.

Household consumption expenditure

Expenditure on savings and investments such as life insurance premiums, pension contributions, debts repaid, interest, litigation, taxes (including road tax, municipal tax etc.), subscriptions to trade unions, ceremonies etc., gifts and charity, remittance to dependents, etc., were excluded from the weighting diagram because they mainly form non-consumption disbursements, and, although expenditure on remittance to dependents, ceremonies, etc., may have some consumption element, they are not priceable since they cannot be identified uniquely with a specific quantity of commodity or service.

Receipts in kind (whether received free or as concessions from employers or others) and gifts received were evaluated at prevailing market prices and the values were recorded as part of income and were also included in the respective expenditure blocks. For self-owned houses (home ownership) and concessional or free houses provided by employers or others, the rental values were imputed on the basis of prevailing rents in the locality for similar rented houses. No distinction was made between the treatment of durable and non-durable goods. However, since purchases of durable goods were infrequent, expenditure was recorded not only for the reference month but also for the reference year and the average monthly expenditure based on reference year values was used for deriving the weights. There was no difference in the treatment of first- and second-hand purchases. Expenditure for health care consisting of medicines purchased, doctors' services and related medical services and contributions to health schemes was included in the weighting diagram.

Method of data collection

The sample of markets and outlets popular with index families was selected through a market survey conducted during 1983 in each urban centre, using a suitable combination of purposive and random sampling methods.

Prices for all items are collected each month from markets, controlled shops, fair-price shops, consumer co-operative stores (including the super bazaars), other retail outlets and service establishments, by the Field Operations Division of the National Sample Survey Organisation, a permanent field organisation for conducting large scale surveys and the regular collection of statistical data.

The prices for different commodity groups are collected on fixed days of the week as far as possible so that comparisons between two quotations from the same outlet are not affected by difference in the timing of the collection. The choice of the days on which price data for a particular commodity group are to be collected is made, keeping in view the volume of transactions on different days of the week.

The prices reported are those paid for actual transactions, inclusive of sales tax and other such charges normally payable by the consumer and net of discounts or rebates commonly allowed. Illegal (black-market) prices, hire-purchase and credit terms, second-hand purchases and trade-in of used goods are not included in price collection. Import prices as such are not included in the price collection. They are rarely, if ever, included in the consumption of index families and, in addition, there are no separate outlets where these items are regularly and exclusively available.

Housing

Rent data are collected from a sample of rented dwellings through a continuous house rent survey conducted every six months. All the sample rented dwellings canvassed for the detailed house rent survey as a part of the 1982-83 household expenditure survey constituted the basic framework for the continuous house rent survey. These dwellings were distributed uniformly over the six months, July to December, as follows: dwellings canvassed from July 1982 to January 1983 are assigned to July, those canvassed from August 1982 to February 1983 assigned to August and so on. From the list assigned to a month, a sample of 14 rented dwellings per investigator was drawn after stratifying the dwellings into four strata according to the number of rooms. Data on rents are collected from sample dwellings assigned to different months at intervals of six months. The house-rent index is compiled from the house-rent data for six months to represent the average half-yearly change in house rent.

In addition to rented dwellings, owner-occupied and rent free dwellings are included in the overall house-rent index. The rent index calculated from the rent data for rented dwellings is used as a proxy for the rent index for owner-occupied dwellings.

The average of the rent relatives for the current six months is calculated over the immediately preceding six months for each of the rent quotations and the rent index is computed as a chain index. Average rent relatives are calculated for each stratum and the averages for the four strata are combined using the actual composition of owner-occupied and rented dwellings available from results of the household expenditure survey to derive the overall average rent relatives. The rent index for the rented and owner-occupied components is calculated as a chain index using the rent index for the previous six month period. The rent index for rent-free dwellings is fixed at 100 and the overall house-rent index is calculated on the basis of an index for rent-free dwellings and an index for owner-occupied and rented dwellings using as weights the proportion of households in the respective categories, available from the household expenditure survey.

Specification of varieties

Specifications for various items were established in sufficient detail with respect to quality, variety and other essential physical characteristics such as size, dimensions, material content etc., bearing in mind the prevalent conditions in different centres, to enable unambiguous identification.

The specifications for each commodity popular with index families were determined using both a market and a consumer level approach, and were earmarked for selected outlets for regular price collection.

Substitution, quality change, etc.

Procedures were laid down for cases in which a particular price quotation consisting of specifications of outlets is not available in the month. Substitute price quotations are then collected, either from reserve outlets for the same specifications or for comparable popular specifications, and the new quotations are linked to the old ones through the usual splicing techniques. Quality changes and the disappearance of old products are treated in the same way, to the extent that they reflect a change of specifications.

Seasonal items

Item coverage in the index for fruit and vegetables varies from month to month depending on their availability in the month. The following method is used for assigning weights to such seasonal items in the index: the sub-group weights for fruit and vegetables are calculated on the basis of average monthly expenditure covering all the items in the sub-groups. However, the individual item weights for the priced items in different months are calculated by distributing the weights of unpriced items to those of priced items on a pro-rata basis.

Computation

The index is computed according to the Laspeyres formula as a weighted arithmetic average with fixed base, using weights corresponding to July 1982-June 1983.

The index is computed directly for each of the centres. The All-India index is computed as the weighted average of the centre indices, using weights based on the estimated total household expenditure in the different centres, using the 1981 population census data and the 1982-83 household expenditure survey data.

Other information

Consumer price indices for industrial workers and for agricultural labourers are also computed by the Labour Bureau and published in the Indian Labour Journal.

Organisation and publication

The index is computed by the Central Statistical Organisation, New Delhi and published in: Monthly Abstract of Statistics.

Idem: Statistical Abstract.

Labour Bureau, Simla: Indian Labour Journal.