11.2.1 At the national level, there are four Consumer Price Index (CPI) numbers. These are:
CPI for Industrial Workers (IW),
CPI for Agricultural Labourers (AL), CPI for Rural Labourers (RL) and
CPI for Urban Non-Manual Employees (UNME).
11.2.2 The base years of the current series of CPI(IW), CPI(AL) and CPI(RL), and CPI(UNME) are 1982, 1986-87 and 1984-85, respectively. While the first three are compiled and released by the Labour Bureau in the Ministry of Labour, the fourth one is released by the Central Statistical Organisation in the Ministry of Statistics and Programme Implementation.
CPI for Industrial Workers, CPI(IW).
11.2.3 The Current series of CPI(IW) on base 1982=100 replacing the old series of 1960 base with effect from October, 1988, covers industrial workers employed in any one of the seven sectors namely factories, mines, plantation, railways, public motor transport undertakings, electricity generation and distribution establishments as well as ports and docks. The index covers only manual workers irrespective of their income. The weighting diagram for this index was derived by conducting a Working Class Family Income and Expenditure Survey (WCFIES) in 70 selected centres (see Annexe 11.1) during 1981-82. Centres were selected purposively keeping in view the administrative needs for centre-based indices and adequate representation to geographical areas and industries. The process of selection of centres was done in three stages:
The total number of centres (70) was allocated among the factory, mining and plantation sectors in proportion to the total employment in the country.
The number of centres allocated to each sector were then distributed amongst States on the basis of industrial employment in a State subject to a maximum of five centres in a State in that sector.
The actual centres were then selected on the basis of centre-wise industrial employment in consultation with the respective State Government.
11.2.4 Though, a centre is selected on the basis of workers in 3 sectors namely Factory, Mining and Plantation, but the detailed survey covered all the 7 sectors.
11.2.5 Centre-wise, the item basket was determined on the basis of WCFIES. The items retained in the basket accounted for a substantial share of expenditure in the group or sub-group of items and could also be allotted a price over the life of the series. The all-India index is a weighted average of 70 centres’ indices. The weight assigned to each centre is the proportion of the estimated consumer expenditure of the centre to the aggregate consumer expenditure of all the centres. All-India weights of CPI(IW) at group and sub-group levels are given in Annexe 11.2.
11.2.6 Representative specifications of items for the purpose of pricing were selected by conducting a market survey in each centre. The retail prices used in the compilation of CPI(IW) are collected by the Labour Bureau through part-time price collectors, who are employees of the State Directorates of Economics and Statistics or State Labour Commissioners’ offices on weekly, monthly and half yearly basis and from selected markets and shops. A six-monthly house rent survey is conducted by the staff of the Labour Bureau for collection of house rent data for compilation of house rent index, which is revised in January and July every year. The present CPI(IW) is under revision on the basis of FIES conducted in 78 selected centres during 1999-2000.
CPI for Agricultural Labourers and Rural Labourers, CPI (AL/RL)
11.2.7 A person is treated as an agricultural labourer if he or she follows one or more of the agricultural occupations in the capacity of a labourer on hire, whether paid in cash or kind or partly in cash and partly in kind. A rural labourer is defined as one who does manual work in rural areas in agricultural and non-agricultural occupations in return for wages in cash or kind, or partly in cash and partly in kind.
11.2.8 The source of weights for the current series of CPI(AL) and CPI(RL), with base 1986-87, released since November 1995, is consumption expenditure data collected during the NSS 38th Round of Consumer Expenditure Survey, 1983. For the purpose of collection of consumer expenditure data for deriving weighting diagrams for CPI(AL/RL) as a part of general consumer expenditure survey of NSSO, the rural labour household is defined as one which derives its major income during the last 365 days from wage paid manual employment (rural labour), vis-à-vis wage paid non-manual employment as also self-employment. From amongst the rural labour households, those households which earn 50% or more of their total income (from gainful occupation) during the last 365 days from wage paid manual labour in agriculture are categorised as Agricultural Labour Households. All-India weights of CPI(AL/RL) at group and sub-group levels are given in Annexe 11.2. This series is presently compiled for 20 States and All-India. Monthly price data collected from 600 villages spread over 20 States (see Annexe 11.3) by the field staff of FOD are used in the compilation of these indices. The sample of 600 villages is staggered over four weeks of a month with one-fourth of the sample covered every week. Prices are collected on the fixed price collection day which may be a “Hat” day for “Hat” or non-daily markets and any market day for daily markets. CPI(AL) is a sub-set of CPI(RL) series. The rural retail prices for these two index series are the same but the weighting diagrams are different.
CPI for Urban Non-manual Employees, CPI(UNME)
11.2.9 An urban non-manual employee is defined as one who derives 50 per cent or more of his or her income from gainful employment on non-manual work in the urban non-agricultural sector. The current CPI(UNME) series with base 1984-85, introduced in November 1987, derives the weighting pattern from the family living survey conducted during 1982-83 in 59 selected urban centres. All-India weights of CPI(UNME) at group and sub-group levels are given in Annexe 11.2. Centres were selected keeping in view, (a) concentration of the UNME population at the centre, (b) inclusion of State capital cities, and (c) regional representation. The centres were allocated to different States broadly in proportion to their 1981 urban population with the limitation that not more than five centres were allotted to any State. For compilation of CPI(UNME), retail prices in respect of the selected items and services from selected markets in the 59 centres are collected by FOD on a monthly basis. The index is being released by CSO with a time lag of about two weeks. Some of the State Governments, public and private sector undertakings, foreign embassies, etc. are making use of this index for purposes of regulating Dearness Allowance. The index is also used for computing the advance tax liability of tax payers from capital gains by the Central Board of Direct Taxes.
11.2.10 CPI numbers presently compiled relate to different base years and cater to specific segments of the population, and thus can be considered as partial indices. These indices are not oriented to reflect a true picture of the price behaviour and effect of price fluctuations of various goods and services consumed by the population in the country, over a period of time.
11.2.11 The Ministry of Statistics and Programme Implementation has proposed to undertake a compilation of CPI for entire rural and urban areas separately as a Plan activity. However, most of the revision exercises of CPIs show that it is difficult to obtain the necessary complement of resources including manpower even after the clearance of the scheme at the highest level in the Expenditure Finance Committee (EFC). For example, the scheme on CPI(IW) was cleared by the EFC in 1996. But the work could start only in 1999 due to various administrative reasons. To overcome these difficulties and to complete the revision exercise as per schedule, there is a need to streamline the administrative procedures so as to get the requisite resources such as hardware and manpower, etc. in a time-bound manner, once the scheme is cleared by the Expenditure Finance Committee (EFC).
11.2.12 For compilation of All-India CPI(IW), the difficulty arises mainly in demarcating the geographical regions which the different centre-indices can be taken to represent and also in deriving estimates of the total number of working class families in these regions. This is largely due to the fact that the choice of the constituent centres was not determined by the laws of probability but was made purposively (in consultation with the various State Governments) with a view to representing as many of the industrially important centres in each state as possible subject to the total number of centres being limited to 70. Therefore, adequate representation to all the States could not be ensured. The index of each of the selected centres in a State has, therefore, to be assumed to represent a larger section of the State’s industrial population than what it might actually be able to. Problems in relation to the present CPI(IW) series, are:
The Technical Advisory Committee on Statistics of Prices and Cost of Living Index Numbers (TAC on SPCL) felt that large cities, where the price situation would usually be subject to peculiar stresses and strains, should not be over-weighted in respect of their contribution to the all-India Index. Hence the TAC on SPCL recommended a method of allocation according to which each centre is assumed to represent an equal share of the working class population within the State, unless the actual working class population for a particular centre itself exceeded the proportionate share accruing to the centres was taken to determine its share and the remainder of the working class population in the State as a whole was equally distributed among the remaining constituent centres in the State.
Precise figures relating to the number of working class families are not available for the different States. Firstly, an average at the State-level for the figure of ‘number of earners per family is worked out by taking a weighted average of corresponding figures for the constituent centres within the States, the weights being the estimates of the number of working class families relating to seven sectors at the centre. The population of working class families for a State was obtained by dividing the average daily employment in respect of the sectors covered by the average number of earners in the State. Thus it is seen that a number of assumptions seem to be inescapable in the process of arriving at the final weighting diagram for the all-India index.
The coverage of the working class population for the all-India index falls short of the entire Indian Union to the extent of exclusion of a few States and Union Territories, like Himachal Pradesh, Tripura, Goa, Daman and Diu, Andaman and Nicobar Islands, Lakshadweep, Meghalaya, Manipur, Nagaland, Mizoram, Arunachal Pradesh, Sikkim, Dadra and Nagar Haveli.
11.2.13 In 1988, the Working Group on Price Statistics after considering the various aspects of collection of rural retail prices for the index including the present sampling frame of markets and price villages recommended that the overall sample size of villages be enhanced from 600 to 800 at the all-India level. The NSSO considered the recommendations of the Group and suggested some modifications in the procedure to be followed for collection of rural retail prices, keeping in view the availability of resources. The modified scheme envisaged collection of prices initially from a new set of 600 villages and subsequently, from 1000 villages with replacement of one-fifth of the sample villages every year. The Commission observed that the number of villages from which the rural retail prices are being collected still continued to be 600 only.
Conclusions and Recommendations.
11.2.14 All Consumer Price Indices (CPI) presently compiled, are based on outdated base years. With a long period having elapsed since considerable changes in the economy have taken place affecting the general patterns and levels of consumption. Moreover, several new commodities and specifications have come into vogue while some of the earlier items and specifications have disappeared altogether from the market. The frame of outlets and rented dwellings prepared earlier may not be of any use now for substitution purposes. The Commission, therefore, recommends that:
It should be mandatory on Labour Bureau and CSO to revise their series preferably every five years, but not later than ten years. In the case of CPI (AL) and CPI (RL), the base year should be revised five yearly, using the quinquennial NSS Consumer Expenditure Survey data.
For revision of the series, the administrative procedure should be streamlined in such a way that necessary resources are sanctioned in a time-bound manner. The exercise on revision of the base year should be completed by the concerned agencies within a one- year period preferably after the availability of the requisite data.
The procedure of allocation of weights to different centres covered under CPI(IW) for computation of the all-India CPI(IW) needs further examination by TAC on SPCL. A similar examination should also be done for CPI(UNME) series.
Increasing the present sample of 600 villages spread over 20 States selected for price collection under CPI (AL) and CPI(RL) should be examined by TAC on SPCL.
As the current CPI series does not provide changes in the prices for the entire rural and urban population since they are designed to measure the changes in the prices of goods and services consumed by specific segments of the population, there is a need to compile the CPI separately for the entire rural and urban populations. TAC on SPCL should give a methodology for compilation of CPI for rural and urban areas separately using quinquennial NSS Consumer Expenditure Survey Data for the preparation of the weighting diagram. TAC should also give a procedure for compiling a combined index based on these two indices. The existing system of price data collection should be suitably streamlined and augmented so as to provide price data for compilation of CPI for rural and urban areas.