Indian Wholesale and Consumer Price Indices

Brian Sturgess - September 2011

1. Indian Price Indices

The official price indices compiled and published in India are based upon two types of price data: wholesale and retail transactions. After years of planning a new all India Consumer Price Index (CPI) for all categories of both urban and rural working households was launched with effect from February 2011 which will start producing consumer price inflation estimates from January 2012. This is produced by the Central Statistics Office of the Government of India. [1] Prior to the launch of the new CPI four different types of consumer price indices were produced. These indices aimed at measuring the price level of different baskets of consumer goods purchased by specific types of workers: urban and rural; industrial and non-manual. The Central Statistical Organisation published one of these four price indices that for the middle-classes or urban non-manual employees (CPI-UNME), while the three other consumer price indices for industrial workers (CPI-IW), for agricultural and for rural labourers (CPI-AL and CPI-RL) are all compiled and published monthly by the Labour Bureau of the Government of India.[2]

The Wholesale Price Index (WPI) is based on the weighted average price of a basket of a large number and variety of industrial and farm transactions and is published every month. A number of WPI sub-indices are also published for both broad and narrow sub-categories of goods included in the WPI. The WPI for all goods covered is published by the Office of Economic Advisor, Ministry and Commerce (OEA) every month. For some WPI categories index price level data are made available on a weekly basis. [3] Consequently, the WPI has become the most watched headline measure of price inflation in India.

Prices for industrial products and natural resources are also monitored by various Indian Ministries and Government Departments for the sectors within their domain including chemicals and petrochemicals, natural gas. In addition, the Department of Consumer Affairs [4] also publishes data on the prices of 14 commodities deemed to be essential from the point of view of their impact on the cost of living of the most vulnerable sections of the country’s population.   Indian price data is also available from the calculation of measures of the economy wide price level through the indices used in the National Accounts Statistics to deflate GDP and Private Final Consumption Expenditure, but these are published quarterly.

In recent years there has been much debate among statisticians and economists in India about the extent that the price indices available are adequate for the uses to which they are put in particular measuring inflation and the cost of living of Indian households. Until relatively recently there have only been piecemeal changes in the system of measuring price indices in India despite the findings of the National Statistical Commission which reported on the main problems in 2001 after critically reviewing all of the available price indices in India and finding deficiencies in systems of price data collection.[5] However, the last two years have seen greater urgency in the process of revising the methods by which price indices are measured in India culminating in the discontinuation of some indices, major overhauls in the methods of construction of others and the launch of the new all India CPI. This paper will review the main price indices published in India highlighting some of their main statistical and economic strengths and weaknesses and will discuss both the proposed and the actual changes that have been made.

2. Consumer Price Indices

The New All India CPI: There have been a number of changes made to try and improve the quality of consumer price indices published in India in recent years. The most important has been the launch of the new All India CPI introduced after an initiative of the Reserve Bank of India. The Central Statistics Office (CSO), Ministry of Statistics and Programme Implementation subsequently launched a new series of Consumer Price Indices (CPI) (Base Year 2010=100) for all-India for Urban, Rural and Urban and Rural combined with the first index calculated for January 2011. The new price indices are compiled at State, Union Territory and at all-India levels[6] and are published monthly, but usable annual consumer price inflation data will not become available until January 2012. 

The category weights for the new series have been based on the average monthly consumer expenditure of urban and rural households according to the NSSO 61st round Consumer Expenditure Survey data for 2004-05. All surveyed consumption expenditure was classified into consumption groups and subgroups on the basis of the Classification of Individual Consumption according to Purpose (COICOP) the international standard adopted by the United Nations. [7] The category weights adopted by the new CPI are shown in Table 1 and can be contrasted with the weights of the other four other CPIs discussed below which have been used in India. Apart from the weights used in the CPI-IW for industrial workers the weights of the other price indices have not been updated for several decades. These weights are given in Table 2.  Multiple norms were used to select items for inclusion in the new CPI. These were to include all Public Distribution System (PDS) items[8] and other criteria based upon the significance of the item. [9]

Housing expenditure is based on the compilation of a house rent index based on a rent equivalence approach for self-owned dwellings and rent paid for tenant properties. Housing expenditure was found to be negligible and was omitted for rural households, but has a weight of 9.77% for the new combined all India CPI. In contrast to the older urban based CPI-UNME for middle-class households and the CPI-IW for industrial workers, the weight of housing expenditure rises more realistically in the new CPI for urban households to 22.53% compared to 16.41% and 8.67% respectively. Housing expenditure had also been omitted in the two rural consumer price indices compiled the CPI-AL and the CPI-RL for agricultural and rural labourers.

The proportion of consumer expenditure allocated to food, beverages and tobacco at 59.31% for Rural households (CPI-AL: 72.94% and CPI-RL: 70.4%) and 37.15% for Urban Households (CPI-IW: 60.15% and CPI-UNME: 47.13%) and 49.71% for combined rural/urban households differs significantly in the new CPI from the other four older consumer price indices. Consumer expenditure weights for Fuel and Light for rural households in the new CPI rise to 10.42% compared to 8.35% for the CPI-AL and 7.90% for the CPI-RL. The weights for Fuel and Light at 8.40% for urban and 9.49% for combined households is also raised on the figures for the other urban consumer price indices - 5.48% for the middle-class CPI-UNME and 6.28% for the CPI-IW for industrial workers.

The Miscellaneous category of goods shows some significant revisions for the weights for consumer expenditure on services such as education and medical care. A combined weight of 5.69% of consumer expenditure is allocated to medical care in the new CPI broken down into 6.72% for rural households and 4.34% for urban ones. This compares with 2.51% for the urban CPI-UNME and 2.59% for the CPI-IW while the rural indices the CPI-AL and CPI-RL allocated weights of 4.38% and 4.23% respectively for this category. In the new CPI education is given a weight of 4.18% for urban households and 3.35% for combined rural and urban. In contrast, the older CPI-UNME, representing the expenditure of middle-classes, and the CPI-IW, representing industrial workers, had weights for education of 4.58% and 1.74% respectively. In the new CPI rural households were given a weight of 2.71% for education while the CPI-AL allocated a weight of only 0.41% and 0.39% for the CPI-RL.


New All India CPI Consumer Expenditure Weights

Category Weights



(Rural and Urban)

Food/Beverages and Tobacco




Fuel and Light




Clothing, bedding and Footwear
















Source: Central Statistics Office

The new CPI has a broader geographical and commodity coverage than the other consumer price indices discussed below. The CPI-AL and the CPI-RL compiled by the Labour Bureau used retail prices collected by the NSSO from fixed markets in 600 sample villages in 20 states. The new CPI Rural has expanded coverage up to a total of 1181 villages. The new CPI Urban collects data from 310 towns with 1114 price quotations covering all Indian states and Union Territories. This compares with 59 urban centres for the now discontinued CPI-UNME and 76 for the CPI-IW. The new CPI numbers will be revised on the basis of the results of the next round of the Consumer Expenditure Survey by the National Survey Sample Organisation (NSSO)[10] of the Ministry of Statistics. This is scheduled to be conducted during 2011-12 and revisions will be undertaken every five years thereafter.

Category Based CPIs: India has used four separate consumer price indices focussed on categories of workers. These consumer price indices were all regularly calculated until 2010 for the cost of a basket of consumer goods faced by a reference population. The Indian Labour Bureau compiled an index for Industrial Workers (CPI-IW) which measures the change in the retail prices of fixed basket of goods and services being consumed by an average working class family. The Bureau also published a monthly price index for two other target groups: agricultural labourers (CPI-AL), and rural labourers (CPI-RL). The reference group for the fourth index was middle class urban families or urban non-manual employees (CPI-UNME) and this price index has until recently been compiled and released by the Central Statistical Organisation (CSO) of the Ministry of Statistics. The main features of these four consumer price indices are shown in Table 2.


Household Category CPI Consumer Expenditure Weights






Base Year





No of items in basket





Collection Points





Time lag

8 weeks

4 weeks

3 weeks

3 weeks

Weight of Food, Beverages and Tobacco





Weight of Fuel & Light





Weight of Housing





Weight of Clothing and Footwear















          Source: Sharma (2010)

Middle Classes (CPI-UNME): The CPI-UNME has been compiled by the Central Statistical Organisation on a monthly basis from 1961. Price collection for this index was discontinued in April 2008. The most recent base year for 1984-85 stems from the findings of a middle-class family living survey covering 45,000 UNME families in 59 selected urban areas which was conducted during 1982-83. Using the weights determined by this survey the retail prices of goods and services were collected monthly by the National Survey Sample Organisation (NSSO)[11] of the Ministry of Statistics in 1022 markets in the 59 urban centres. The number of items for which prices were surveyed varied between centres with the smallest number in Imphal (146) and the largest number collected in Delhi (345).

A meeting of the National Statistical Commission in 2008 decided to discontinue the collection of price data for the CPI-UNME from April of that year and to issue a Linked all India CPI-UNME price data based on the CPI-IW until the new CPI-Urban was released. This was finally accomplished in January 2011 as discussed above and consequently calculation of the Linked all India CPI-UNME was terminated at that time.[12] The CPI-UNME has been calculated using a Laspeyres Index formula and data from the most recent base year 1984-85 were released in November 1987 providing a continuous series until its recent termination.

Industrial Workers (CPI-IW): This consumer price index is compiled by the Labour Bureau of the Ministry of Labour using retail prices collected from 261 markets in 76 centres by officials of State Governments and Labour Commissioners. The earliest base year 1960 was based on an expenditure survey carried out into the living standards of working-class families in 1958-59. [13] An updated working class family survey carried out in 1981-82 resulted in an updating of the base year to 1982 which was refreshed again and the base year of the current series of the index is 2001. The weights in the current base year arise from a working class family expenditure survey carried out in 1999-2000. [14] The number of items in a consumer basket surveyed varies between 120 and 160 depending on the consumption pattern of each centre and other administrative factors. The index numbers are compiled using a Laspeyres index formula. 

Agricultural Labourers (CPI-AL) and Rural Labourers (CPI-RL): The Labour Bureau also compiles consumer price indices for Agricultural Labourers and Rural Labourers on a monthly basis. An agricultural or a rural labour household is considered as one deriving 50% or more of its income from the employment of one or more of its members from agricultural or rural labour. These persons are considered to be involved in agricultural labour if they follow a specified list of occupations in the capacity of receiving paid wages whether remuneration is in cash or in kind. [15] Rural labour households are the broader category of worker and include agricultural labourers as defined above and all other rural labourers. The current base year of both series is 1986-87 and the weights for both are based on estimates generated fro data collected through a consumer expenditure survey conducted by the NSSO in its 38th round in 1983. The NSSO collects retail price data for 260 items of goods and services in 20 states from fixed markets in 600 villages. These indices are calculated using a Laspeyres weighted aggregate method to arrive at an all India index. The method is described in Sharma (2010).

Evaluation: The National Statistical Commission reporting in 2001 criticised India’s four CPIs on a number of grounds including a lack of uniformity between indices in the base year chosen, in the weights and the long lags in updating them as well as some technical issues relating to sample selection. The data in Table 2 shows that the four indices differ significantly in geographical coverage and in their weights. The CPI-UNME classifies goods and services consumed into five main groups while the CPI-IW uses six. In contrast both the CPI-AL and the CPI-RL employ only four categories. The expenditure surveys found that the housing cost of the rural population was considered to be negligible in the base year and so it has been omitted as a category in both price indices. The indices were all developed separately to meet different policy requirements and although substantial differences would be expected between rural and urban households, the inconsistencies in different surveys means that the four consumer price indices cannot be used to reflect price behaviour in the Indian economy.  Consequently, the National Statistical Commission Report of 2001 recommended the development of all India consumer price index compiled separately for urban and rural households and combined after appropriate and consistent weightings have been applied.

There are other shortcomings of the four indices of which the long time lag between surveys and base years is serious. For example, Table 2 shows that three of the CPIs employed base years that were based on surveys over a quarter of a century old. This ignores the significant changes that would have been expected to have taken place in Indian consumption patterns arising from structural change, demographic movements and changes in preferences and spending patterns occasioned by trade and rising disposable income. [16] Another problem is the omission of the informal sector of the Indian economy which accounts for a significant amount of the Indian economy. One official estimate placed the informal sector at 55% of Indian GDP in 2006 [17] and it has been suggested by Sharma (2010) that there is a need for a separate CPI series for rural and urban poor covering the lowest three or four deciles in the distribution of income.

3. Wholesale Price Index (WPI)

The WPI is published by the Office of Economic Advisor, Ministry and Commerce (OEA) [18] and is based on regularly measuring several thousand price quotations for items in a basket of 1,224 manufactured goods and primary commodities. The WPI is distinct from the CPI measures of inflation in three main ways: the range of goods included in the basket, the weighting system which is based on industrial and agricultural rather consumer expenditure patterns and the emphasis on bulk rather than on single good purchases.

The WPI for all items is constructed from a weighted price data for three major groups of products. There are Primary Articles (Food, Non-Food and Minerals) with 105 items; Fuel, Power, Light and Lubricants with 19 items and Manufactured Products with 1,110 items in the basket.[19] Price data is collected from a variety of official sources and non-official sources including chambers of commerce, trade associations and leading manufacturers. A Laspeyres base-weighted formula is used to compile the WPI. Commodity and sub-group and group indices are based on simple weighted averages. The current WPI series has the year 2004-5 as a base and the main categories and weights are shown in Table 3.

The latest series of the WPI provides consistent price data from April 2005 and monthly measured inflation figures are shown in Figure 1 from that date. For all uses other than for academic research purposes, the OEA warns that data should only be used from August 2010, with the figures first released on September 14, 2010. The practice of publishing the WPI for all commodities every week was discontinued from October 2009, but prices are still published weekly for primary articles and fuel.  These figures are published with a time lag of two weeks with a revision of the provisional figures made after eight weeks. However, despite the relative timeliness of the release of WPI figures compared to CPI price data its use is limited by the often substantial revisions made later to provisional publication.


Wholesale Price Index Weight Changes by Base Year

Major Group/Group

Weights Base 1981-82 %

Weights 1993-94 %

Weights 2004-05 %

Change 2004/05 to 1981-82

I. Primary Articles





(a) Food Articles





(b) Non-Food Articles





(c) Minerals





II. Fuel, Power, Light and Lubricants





III. Manufactured Products





(a)   Food Products





(b)   Beverages, Tobacco, & Tobacco Products





(c)   Textiles





(d)   Leather & Leather Products





(e)   Others










Source: Sharma (2010) and Office of Economic Advisor

Evaluation: One basic limitation with the WPI is that the concept of a wholesale price is not clearly defined due to the voluntary nature of the data capture process. Prices measured and included in calculating the index can refer to inflationary pressures at varying stages in the supply chain of commodities from primary production to almost final consumption. Competitive conditions across markets will determine the extent to which changes in the WPI will be correlated with changes in more orthodox yardsticks of price inflation. A paper presented at a staff seminar at the Indian central bank last summer notes the qualification made by the National Statistical Commission on the loose definition of wholesale prices:

“…owing to the wide variety of sources, centres, and specifications and due to the practical compulsion of collecting data by the voluntary method, it is difficult to maintain uniformity in the concept of wholesale prices in the collection of price data. In many cases, these prices correspond to farm-gate, factory-gate or mine-head prices; and in many other cases they refer to prices at the level of primary markets, secondary markets or other wholesale or retail markets.” [20]

Another problem with the WPI has been its volatility in recent years. The chart of monthly WPI for all commodities illustrated in Figure 1 shows that wholesale prices in India were rising at an average rate of around 7% per year from April 2005 for twelve months before falling to an average annual rate of 3.2% in the month of October 2006. This was followed by another climb in the WPI to record an annual average monthly inflation rate of 11.1% by July 2008 before deflationary pressures pushed the annual change in the WPI of minus 0.4% in June 2009 prior to another steep climb up to an inflation rate of 10.9% by April 2010. Since then inflation in the Indian economy measured by changes in the WPI has stayed high averaging 9.3% between May 2010 and April 2011.


            Source: Office of Economic Advisor.

Figure 1 illustrates that the WPI category Food is far more volatile than WPI Manufacturing as is the category WPI Fuel and Power. The sizeable weightings of these two sectors in the WPI – 14.337% and 14.991% respectively explains some of the enhanced volatility of the WPI in recent years as crude oil and food prices have moved up and down significantly. This illustrates a general point that the weights assigned to products and product categories can have a significant impact on final estimates of inflation in any aggregate index.

The optimal choice of category weights is more problematic for a wholesale price index than a consumer index. While the weights of items in consumer price indices are based upon surveys of the spending patterns of representative households, the concept of a representative farm, enterprise or distribution is far more nebulous. The WPI weights are based on the relative importance of transactions in the economy as with consumer price indices, but the weights are calculated not as a proportion of value added to GDP, but on the relative value of the commodity’s turnover. This method can lead to distortions and an overemphasis of the importance of some sectors compared with others. The assigned weights of many commodities and groups of commodities in the WPI are not always proportionate to the relative share of the trade of these commodities or groups of commodities in GDP. For example, an economist at the Office of Economic Advisor in a recent paper has estimated that while the ratio of value added to the value of output is around 85% in primary sectors such as agriculture and mining in manufacturing it is around 21%.[21]

One of the most important criticisms of the WPI, however, is that it excludes two important categories of value-added in GDP – the informal sector- which is difficult to measure, but also the WPI does not include any price data for services which are a rapidly growing sector of the Indian economy. The National Statistical Commission concluded

“On the issue of compilation of the Wholesale Price Index (WPI), the Commission has recommended that the revision of base year must be undertaken more frequently so as to capture the changes in industrial structure on account of liberalisation. The Services Sector has presently developed to such an extent now that it contributes significantly to country’s GDP. On account of the non-inclusion of Services Sector in the existing WPI, the development of a separate price index for Services Sector is recommended. Ultimately, this should be merged with the WPI, once the Services Sector index is stabilised and its robustness is established.”[22]

Recent Improvements: The National Statistical Commission Report of 2001 recommended that in order to capture changes in the economic structure of the Indian economy that it was necessary to revise WPI base every five years and to increase coverage. The most recent change in the WPI was to update the base year from 1993-94 to 2004-05 as a result of the recommendations of a WPI Working Group chaired by Professor A. Sen[23]  This also lead to an increase in the coverage of commodities from 435 to 1,224 with the number of commodities in the manufacturing sector rising from 318 to 1,100. The number of commodities in fuel and power remained unchanged at 19 while the number of primary articles rose from 98 to 105.

In consideration of changes in the structure of the Indian economy away from agricultural and extractive towards industrial production the weight allocated to primary articles has fallen from 32.295% for the WPI base 1981-82 to 20.118% in the latest edition with 2004-05. The weight of Fuel, Power, Light and Lubricants rose over the same period from 10.633% of the WPI to 14.910%. The most significant increase was in the category of Other Manufactured Products whose weight rose from 32.187% to 45.075% reflecting the large increase in the number of industrial prices monitored. The importance of more traditional industries, such as manufactured Food Products, Beverages, Tobacco & Tobacco Products, Textiles and Leather and Leather Products, have actually fallen in the composition of the WPI over time. These changes are shown in more detail in Table 3.

The WPI still excludes services, but an Expert Group chaired by Professor C.P Chandrasekhar is currently working on the construction of a Services Price Index (SPI) which will include such items as the price of taxi fares, air tickets, banking fees, insurance premia and postal services with a base of January 2008. It is planned eventually to merge the SPI with the WPI.

4. Divergences between Indices

There is some evidence that over long periods measures of Indian inflation have moved closely together after leads and lags have had time to work themselves out. Whereas it might be expected that the four consumer price indices might diverge to some extent given the different sectors of the Indian population that they measure, the evidence is that there is a significant degree of co-integration between them. A similar long-term pattern is shown between other measures of inflation in India. This is shown in Table 4 which shows annual average inflation rates over a number of decades measured by the WPI, the CPI-IW, the GDP Deflator and the PFCE Deflator. However, it is also evident from Figure 2 that despite these long-term relationships measurements of inflation by the WPI has been diverging from the CPI-IW in recent years.[24]


Annual Average Inflation over Decades

Decades % per annum



GDP Deflator

PFCE Deflator

1971-72 to 1980-81





1981-82 to 1990-91





1991-92 to 2000-01





2001-02 to 2008-09





1971-72 to 2008-09





Source: Mohanty (2010)

Chart 2 shows that the relationship between the WPI and the CPI-IW was fairly close between January 2006 and June 2007.  The difference between the two is measured by the CPI/WPI ration which averaged 100 no real difference between the two indices although in some months the CPI-IW slightly underestimated the price level measured by the WPI. However, after July 2007 the two indices have diverged and the average ratio of the CPI/WPI index was 106 between July 2007 and April 2011.


                 Source: Labour Bureau, Office of Economic Advisor.

Similar divergences between the WPI and other measures of inflation based on changes in price level of goods related to the cost of living have been found by other authors. For example, Singh (2010) found that over the period 2004 to July 2009 the WPI underestimated inflation measured by a comparable Retail Price Index for the important category of commodities monitored by the Indian Department of Consumer Affairs (DCA).[25]  The WPI for these commodities increased to 136.9 by July 2009 producing an average monthly increase of 0.45% (or an annualised inflation rate of 5.37%) while the constructed RPI rose to 144.1 by July 2009 producing an annualised inflation rate of 6.71%. However, the relatively small difference between these average figures disguises a drift between the two indices. An index of the RPI relative to the WPI rose to 105.3 by July 2009 showing an average monthly rate of divergence of 0.11% between the two indices, but the extent of the divergence is exacerbated by the fact that the two measures of inflation stayed relatively close until June 2007 and then widened. RPI inflation for the period April-June 2008 was up by 12.2% on the same period the previous year while the WPI registered a rise of only 7.3%. In the following year the gap closed and the divergence reversed with a rise in the WPI of 13.7% in April to June 2009 compared with an annual rise of 13.0% in the RPI.

The WPI and the CPIs differ in a number of ways apart from in the base years chosen. The two most significant are in coverage and in the category weights chosen. These two differences are discussed below.

Coverage: The WPI covers the prices paid for bulk transactions while the CPIs are based upon the prices paid by different categories of consumers. However, given that the retail market ultimately receives commodities from the wholesale market it might be expected that over time changes in the WPI should be transmitted into the CPIs. Mohanty (2010) looked at time series data covering the period 1973 to 2009 and noted that there was a long-run relationship between the WPI and the CPI-IW indicating at the trend level the CPI lagged behind the CPI-IW by a month on average. [26] Nevertheless, the most recent data shows a widening gap between the CPI-IW and the WPI.

Another important difference in coverage between the indices is that whereas the WPI excludes all services with products covered including only primary commodities or secondary manufactured items, the CPIs include some categories of services albeit to differing degrees. For example, the CPI includes categories such as housing, medical care and education, recreation and amusement. The importance of the services sector in Indian GDP has increased dramatically rising from around 30% of the Indian economy to nearly 60% by 2010. [27] Therefore, changes in the price of services will have a greater influence on CPI measures of inflation than on the WPI.  Between January 2006 and December 2001, the aggregate price of housing in the CPI-IW rose by 63% compared with a rise of 55% in the general CPI-WPI index and only 39% in the WPI.


[5] The National Statistical Commission was set up under the chairmanship of Dr. C Rangarajan to review the system of

collecting economic statistics in India including data on prices and the calculation of price indices.  The 2001 Report can

 be accessed at:

[6] India is a federal country comprising 28 states and 7 Union Territories.

[8] The Public Distribution System (PDS) under the Ministry of Consumer Affairs, Food and Public Distribution is managed jointly with state governments in India and distributes subsidized food and non-food items to India’s poor. The commodities distributed include staple food grains such as wheat, rice, sugar and kerosene at differential rates through a network of nearly 500,000 Fair Price Shops (FPS) across the country to measured Above the Poverty Line (APL) and Below the Poverty Line (BPL) households. 

[9] These criteria are to include all items accounting for 1% or more of total expenditure at sub-group level; to include all items accounting for more than a specified percentage of total expenditure of all consumption items and to include all items for which more 75% of households have reported consumption.

[12]  The link indexed was based on a ratio method after aggregating the sub group

level indices of  Labour Bureau’s  CPI (Industrial Workers) using CPI  (UNME) weights at group/sub-group level for all India. Thiese linked CPI(UNME)  numbers were to be compiled untill the new series of  CPI(Urban) was brought out. Based on the methodology given by National Statistical Commission, a ratio of CPI - UNME numbers to CPI –IW  numbers for each of the 24 months for the two year period from January, 2006, December, 2007 at sub-group level was claculatedt. An average ratio at each level based on 24 months figure was taken as the linking factor at the respective level.  The CPI (UNME) weights were used to get the indices at group and all groups levels.

[13] A working class family is defined as one located in the centre with at least one family member working as a manual worker in one of seven categories of employment (factories, mines, plantations, ports and docks, public motor transport undertakings, electricity generating and distributing establishments, and railways) and which derived 50% or more of its income during the calendar month preceding the day of inquiry from manual work.

[15] These are farming, dairy farming, production of any horticultural commodities, raising livestock, bees or poultry any other activity carried out on a farm.

[17] See Raveendran (2006)

[19] The WPI is based on a weighted arithmetic mean calculated according to the Lasperyre’s index formula.

[20] Nadhanael and Pattanik (2010), p14.

[21] Singh (2010).

[23] Chairman Professor Abhijit Sen:

[24] Mohanty (2010)

[25] Singh (2010’) converted the data collected by the Department of Consumer Affairs (DCA) into an index form using 2004-05 as a base in order to test its relationship with the latest configuration of the WPI. The DCA data is extensive and has monitored the prices of 14 commodities on a daily basis in 17 locations across India. The author constructed a Retail Price Index (RPI) using equal weights for all locations and used the same commodity weights for the 14 goods included in a matched WPI commodity subset: food grains (including wheat flour) – 30.6%; edible oils -10.2% and others 49.2%.

[26] This conclusion was based on Granger Causality tests in a Vector Autoregressive Framework (VAR) using monthly CPI and WPI data. A long-run cointegrating relationship between the CPI-IW and the WPI was also found.

[27] Based on estimates in Eichengreen and Gupta (2010) while Mohanty (2010) estimates services at 65% of Indian GDP in 2010.

Weights: The WPI and the CPIs have very different weighting patterns. The most significant categories are food and fuel. The category of expenditure Food, Beverages and Tobacco has always had a larger weighting in the CPIs than in the WPI. In the most recent version of the WPI (Base 2004-05), food, drink and tobacco has a combined weight of 24.31% including manufactured food products (9.97%). In contrast, food has had a much larger weight in the CPIs ranging from 47.13% in the CPI-UNME to 72.94% in the CPI-AL. In the current version of the CPI-IW (Base 2001) the category has a weight of 60.15%. The differences between the WPI and consumer price indices that are due to differences in the weighting of food are likely to persist despite the launch of the new All India CPI. In the new CPI-Rural the category still accounts for 59.31% while in the CPI Urban it is 37.15% and 49.71% in the combined All India CPI.



Sharma, G.D. (2000) Compilation of Consumer Price Indices (CPI) and

Wholesale price Index (WPI) in India,  Paper presented at Joint OECD/ESCAP Workshop on Key Economic Indicators, Bangkok, 22-25 May 2000

Raveendran, G. (2005), “Estimating Employment in the Informal Economy through Labour  Force Surveys: An Indian Attempt”, Report of the Eighth Meeting of the Expert Group on Informal Sector Statistics. (Delhi Group) hosted by the Fiji Islands Bureaus of Statistics in Sheraton Resorts at Nadi  - 29-31 March, 2005.

Nadhanael, G.V. and Pattanik, S. (2010) RBI Staff Study - 5/2010 Measurement of Inflation in India : Issues and Associated Challenges for the Conduct of Monetary Policy

Mohanty, D (2010) Measures of inflation in India – Issues and

Perspectives,  Speech by Mr Deepak Mohanty, Executive Director of the Reserve Bank of India, at the Conference of Indian Association for Research in National Income and Wealth (IARNIW) at the Centre for Development Studies (CDS), Thiruvananthapuram, 9 January 2010.

Eichengreen, B. and Gupta, P. Mohanty, D (2010) The Service Sector as India’s Road to

Economic Growth? Indian Council for Research on International Economic Relations, Working Paper No. 249, April 2010.