Industrial Production Statistics of India

Special Statistics

I

THE COVERAGE

The objective of this note is to present a set of basic statistics on the physical output of industrial products that comprise the Central Statistical Organisation’s (CSO) indices of industrial production with base years 1980-81 and 1993-94. The index with 1980-81 as the base year covered 352 items, con­sisting of 290 items from the manufacturing sector with a weight of 77.11 per cent, 61 items from mining and quarrying with a weight of 11.46 per cent and electricity as a single item carrying a weight of 11.43 per cent. Whereas the revised index with the base year 1993-94 covers 543 items clubbed into 287 item groups viz., manufacturing (285), mining & quarrying (1) and electricity (1). We have endea­voured to present physical production data for the past 17 years from 1986-87 to 2002-03 in respect of almost all of these items. In order to solve the problem of non-comparability of data with the reorganization of item groups, they are presented in two separate tables with comparable series viz., Appendix IV (A) and Appendix IV (B). Appendix IV (A) provides continuous data for 17 years with respect to the items listed under IIP series with 1980-81 as base. But data are not available in respect of many items from 1997-98 onwards with induction of revised IIP series (1993-94=100). Appendix IV (B) contains the latest data for 9 years from 1994-95 to 2002-03 covering 301 items as included under the latest IIP series with base 1993-94.

These data series have had to be culled out and pieced together from diverse sources. In this regard, the CSO’s basic publication Monthly Production of Selected Industries in India, which was providing production data for all the 352 items in the 1980-81 series, has been dis­continued since July 1995. Now, the output data for 303 (including manufacturing, mining & quarrying and electricity) items, as per the modified list, are being published by the CSO in its generic data sources Monthly Abstract of Statistics and Statistical Abstract India. Simul­taneously, the Data Division of the Depart­ment of Industrial Policy and Promo­tion (DIPP), which is now responsible for the generation of basic data on manufacturing items, publishes on a monthly basis physical pro­duction figures in the Industry Ministry’s new publication SIA Statistics. Initially, the publication provided the production data on 252 items in the non-SSI sector, but from July 1998 onwards, the number of items covered has come down to 213. Further, as per the decision taken by the CSO in October 2000, four items, viz., radio receivers, photosensitized paper, engines and chassis were dropped from the IIP basket. Consequently, from November 2000 onwards, the SIA Statistics has been providing output data on 209 key manufacturing items in the non-SSI sector.

In respect of mining and quarrying, the Indian Bureau of Mines (IBM), Nagpur, has been publishing the quantity and value of output with respect to 61 items in Indian Minerals Yearbook.  From April 1997-98 onwards, the IBM has been providing output and mineral indices data on the website (www.ibm.nic.in) as per the revised series for about 70 items including three minor minerals. Appendix IV (C) presents these minerals statistics. The data with respect to minor minerals are available only in value terms and hence not presented in Appendix IV (C). Apart from this, the aforesaid CSO’s publication, Monthly Abstract of Statistics presents monthly data for 57 “mining and quarrying” items. Together, these two sources provide data for 71 items and five items have been dropped probably because their production figures are irregular.1 For only a few series, these data have had to be supple­mented by using such official publica­tions as the Government of India’s Economic Survey, and the Reserve Bank of India’s Report on Currency and Finance (up to 1997-78). A continuous series of data for 12 years from 1990-91 to 2001-02 are available in respect of 57 mining and quarrying items; the data for the previous years are available for a fewer number of pro­ducts. Data with respect to additional items, as provided by the IBM, are available for only five years from 1997-98 to 2001-02 [Appendix IV (C)]. In addition, the IBM has been bringing out Index of Mineral Production (on its website) from 1994-95 onwards and the same are presented in Appendix II. In respect of coal, which is a single item in the IIP, data for its subdivisions, namely, coking and non-coking coal, and within coking coal, metallurgical and non-metal­lurgical coal, are also included in Appendix IV (C).

For electricity, the data have been obtained from the Central Elec­trical Authority’s Public Electricity Supply: All-India Statis­tics, General Review and also from their website (www.cea.nic.in) from 1997-98 onwards. Similarly, though the IIP combines all sources of power generation into one single item, namely, “electricity”, we have pre­sented production data for (a) utilities and (b) non-utilities, and within utilities, for (i) hydro, (ii) thermal, and (iii) modern sources [Appendix IV (D)]; these data are reproduced from the Government of India’s Economic Survey.

The 18 industrial products, which have output from both small-scale (DCSSI) and medium- and large-scale sectors, are marked bold in Appendices IV (A) & (B), with separate figures presented for these two categories. These 18 items constituted 9.09 per cent of the total weight in 1980-81 series and 5.68 per cent in 1993-94 series. But, unlike the 1980-81 series, the 1993-94 series has assigned specific weights to the DCSSI items.

II

Index of Industrial Production

Index of Industrial Production (IIP) is an abstract number, the magnitude of which represents the status of production in the industrial sector for a given period of time as compared to a reference period of time.

The beginning:

The first official attempt to compute and release the Index of Industrial Production was made by Office of Economic Advisor, Ministry of Commerce and Industry with the base year 1937 covering 15 important industries, which then accounted for more than 90 per cent of the total production of the selected industries. As per the United Nations Statistical Organisation’s recommendation, the general scope of IIP includes mining, manufacturing, construction, electricity and gas sectors. However, the present general index of industrial production compiled in India has mining, manufacturing, and electricity sectors only, due to constraints in data availability on construction and gas sector.

Successive Revisions:

As the structure of the industrial sector changes over time, it becomes necessary to revise the base year of the IIP periodically to capture the changing composition of industrial production and emergence of new products and services so as to measure the real growth of industrial sector (UNSO recommends quinquennial revision of the base year of IIP).  After 1937, the successive revised base years were 1946, 1951, 1956, 1960, 1970, 1980-81 and 1993-94. Initially it was covering 15 industries comprising three broad categories: mining, manufacturing and electricity. The scope of the index was restricted to mining and manufacturing sectors comprising of 20 industries with 35 items, when the base year was shifted to 1946 by Economic Adviser, Ministry of Commerce & Industry and it was called Interim Index of Industrial Production. This index was discontinued in April 1956 due to certain shortcomings and was replaced by the revised index with 1951 as the base year covering 88 items, broadly categorised as mining & quarrying (2), manufacturing (17) and electricity (1) compiled by CSO. The items in this index were classified according to the International Standard Industrial Classification (ISIC) 1948 of all economic activities.

The index was further revised in July 1962 to the base year 1956 as per the recommendations of a working group constituted by the CSO for the purpose and it had covered 201 items, classified according to the Standard Industrial and Occupational Classification of All Economic Activities published by the CSO in 1962. The index with 1960 as the base year was based on regular monthly series for 312 items and annual series for 436 items. Hence, though the published index was based on regular monthly series for 312 items, weights had been assigned for 436 items with a view to using the same set of weights for the regular monthly index as well as the annual index covering the additional items. However, the mineral index prepared by the IBM excluded gold, salt, petroleum and natural gas.

The next revised series of index numbers with 1970 as the base year, had taken into account of the structural changes occurred in industrial activity of the country since 1960 and this index was released in March 1975 covering 352 items comprising mining (61), manufacturing (290) and electricity (1). The working group (set up in 1978) under the Chairmanship of the then Director General of CSO, decided to shift the base to 1980-81, to reflect the changes that had taken place in the industrial structure and to accommodate the items from small-scale sector.

Thus, a notable feature of the revised 1980 index number series was the inclusion of 18 items from the SSI sector, for which the office of the Develop­ment Commissioner of Small-Scale Industries (DCSSI) could ensure regular supply of data. The production data for the small-scale sector were included only from the month of July 1984 onwards; prior to this the production data from the directorate general of technical development (DGTD) for large and medium industries alone had been utilised. For the period April 1981 to June 1984 in respect of these 18 items, average base year (1980-81) produc­tion as obtained from DGTD was utilized. From July 1984 onwards, combined average base year production both for DGTD and DCSSI products was utilised. The weights for these items were based on ASI 1980-81 results and no separate weights for DGTD and DCSSI items were allocated in the 1980-81 series.

The latest series with 1993-94 as the base year containing 543 items (with the addition of 3 items for mining sector and 188 for the manufacturing sector) has come into existence on 27^th May 1998 and ever since, the quick estimates of IIP are being released as per the norms set out for the IMF’s SDDS², with a time lag of six weeks from the reference month. These quick estimates for a given month are revised twice in the subsequent months. To retain the distinctive character and enable the collection of data, the source agencies proposed clubbing of 478 items of the manufacturing sector into 285 item groups and thus making a total of 287 item groups together with one each of electricity and mining & quarrying. The revised series has followed the National Industrial Classification NIC-1987. Another important feature of the latest series is the inclusion of unorganised manufacturing sector (That is, the same 18 SSI products) along with organised sector for the first time in the weighting diagram.

Method:

The index is computed using the weighted arithmetic mean of quantity relatives with weights being allotted to various items in proportion to value added by manufacture in the base year by using Laspeyre's formula:

I = Ʃ (WiRi)/ Ʃ Wi. 

Where I is the index, Ri is the production relative of the ith item for the month in question and Wi is the weight allotted to it.

The Index of Industrial Production (IIP) is a quantitative index, with the production of items being expressed in quantity terms. However, in case of certain items like machinery, machine tools and ship building the unit of reporting is in value terms. To derive the output in real terms, the monthly figure of production value, in such cases is deflated by the Wholesale Price Index (WPI) of the corresponding item/category, released by the Office of the Economic Advisor, Ministry of Industry.

III

Use-Based Classification of Industries

A meaningful classification of appropriate group indices of industrial production relates to what has come to be known as the use-based classification of industries (Appendix II). Pioneering work in regard to alternative classificatory systems was done by the Reserve Bank of India in the early 1970s starting with the 1960 series of index of industrial production [RBI 1970].3 This was subsequently taken over by the CSO, which has been pub­lishing the group indices as given in Appendix III beginning with the 1980-81 series. The latest revision of indices with 1993-94 as the base year, has changed the weighting diagram considerably in case of capital goods (70 per cent fall from 16.4 per cent to 9.7 per cent) and consumer durables (50 per cent rise from 2.6 per cent to 5.1 per cent) sectors.  

The official agencies have by convention adopted the following definition for the use-based classification:

(i)                  Basic goods: Any bulk or raw material products used for further production of new items in agriculture, manufacture or construction.

(ii)                Capital goods: Plants, machinery and goods used for physical capital or assets formation.

(iii)               Intermediate goods: Any product produced as incomplete product or that goes as input at production time for further finishing.

(iv)              Consumer durables: Products directly used by consumers and having signi­ficant durability or resale value (say more than 2-3 years).

(v)                Consumer non-durables: Products, which are directly used by consumers and which cannot be preserved for any significant or long periods (RBI 1970).

All the items (mining and quarrying and electricity as one item each under basic goods besides items of manufacturing) are classified under the aforesaid categories as detailed in Appendix II. Briefly, they are summarised in Table 2 below.

A grey area involved in the above classi­fication has been the advisability of making a distinction between basic and interme­diate goods though basic goods are also of the same genre as intermediate inputs. However, the official agencies have retained the distinc­tion as originally made by the RBI. This clas­si­fica­tion has its own merit because (a) basic goods are essentially bulk products, (b) they originate in primary sources, (c) they are products of infrastructure indus­tries, and (d) above all, they largely constitute finished products used for application elsewhere. On the other hand, intermediate products primarily constitute incomplete products or are by and large intended as intermediate inputs for further produc­tion or processing.

IV

PERFORMANCE OF INDUSTRIAL SECTOR

Vast Fluctuations in Output Growth

Any detailed analysis of industrial statistics as set out here is outside the scope of this note. However, a few highlights of what the data portray as a lead for its users are attempted here. First, an important observation that can be made is that industrial growth has been distinctly unsteady in the post-reform period. This is evident from an optical view of the annual per­centage varia­tions of individual products, which brings out rather strikingly the mind-boggling numbers of negative signs, implying vast year-to-year fluctuations particularly in the period of the 1990s. This is true in case of almost all the items including minor minerals and many chemical items; it has also been true of metals and engineering items.

Second, an overview of industrial production performance during the past three decades (Table 3 & 4) suggests that there has been a slow down in growth in 1990s, after gaining considerable momentum in 1980s. Production of industry sector, as represented by the Index of Industrial Production (IIP), had posted a healthy growth of about 8.0 per cent per annum, driven by a robust growth in manufacturing (7.7 per cent per annum) and electricity (9.1 per cent per annum) sectors, during the 1980s against 4.6 per cent per annum during the 1970s. There was a vast increase in investment and consumer demand in 1980s indicated by a large growth in capital goods sector (11.3 per cent) and consumer durables sector (14.8 per cent). This was reinforced by the performance of two-digit industrial categories with a substantial growth of 20.1 per cent per annum in electrical machinery and a considerable growth in chemicals and chemical products (10.2 per cent), paper and paper products (7.5 per cent), leather and leather products (7.3 per cent), transport equipment (6.7 per cent) and machinery, machine tools and equipment (6.0 per cent). But, during the 1990s, the growth rate of industrial production has decelerated to 6.3 per cent per annum due to slowdown in the growth of all the three major groups. Further, the important growth propellers such as investment and consumer demand - broadly indicated by the performance of capital and consumer goods sectors - have been sluggish. Amongst the two-digit industrial categories, only a few industries including beverages, tobacco and related products (11.1 per cent), non-metallic mineral products (9.7 per cent), transport equipment (9.0 per cent) and wool, silk and man made fibre (8.9 per cent) have posted a moderate growth in the 1990s compared to the 1980s. On the whole, there was a slow down in industrial growth during the latest decade. 

V

Deteriorating Quality of Industrial Statistics

The literature is full of misgivings regarding the quality of industrial statistics in India. The infirmities noticed in this respect relate to both the basic definitional questions relating to coverage of products in industrial statistics and the system of administering the data gathering arrangement. Besides, a distinct fall-out of industrial liberalisation in recent years has been the further deterioration in the quality of industrial statistics.

In order to release IIP within six weeks as per the norms of Special Data Dissemination Standards (SDDS) of IMF, all the 15 source agencies are required to furnish data to CSO within four to five weeks from the close of month. If the regular monthly production data from the unorganised manufacturing sector is not available, then the item basket has been identified on the basis of data from the registered sector only. Even for the registered sector, the quality of production data supplied by the major source agencies suffer from substantial non-response on the part of manufacturing units and consequential estimation resorted to by the source agencies. Further, the source agency responsible for the small-scale sector could not line up the production data for the items of the revised series. As a result, the industrial growth predicted based on the revised IIP does not seem to reflect the ground realities and the quality has to be improved further.

Apart from the fragile basis of the estima­tion of production data collected for small-scale industries (it was initially with 2 per cent sample of the working units but even that sample proportion has declined) as also the extremely limited coverage given to the SSI sector,4 data received by the DGTD – now the Data Division of the DIPP – in respect of all 209 items under manufacturing are from the ‘factories’ falling within the scope of the Industries Development and Regulation Act, 1951 (IDR Act). The scope of the term ‘factory’ as defined in the IDR Act has been very re­strictive as compared with that defined in the Factories Act, 1948. While the former covers establishments engaged in the manufacturing process having 50 workers or more if using power or having 100 workers or more otherwise, the latter has the cut-off point of 10 workers with the aid of power or 20 workers without power. Roughly, about 77 per cent of the factories producing about 15 per cent of ‘factory’ output are by and large excluded from the purview of the monthly production statistics.

Secondly, it has been reported that even before the July 1991 liberalisation of industrial controls, there had been a steady decline in the submission of production returns by indus­trial undertakings. Thus, the units which did not require any special assistance from the government such as, flour milling industries, industrial alcohol, foundries, industrial gases, auto-ancillaries and industrial machinery units, rarely complied with the provisions of IDR Act and submitted the returns [Kulshreshtha and Kolli 1995]. A note prepared by the office of the economic adviser, ministry of industry (1995) for a seminar had the follow­ing observations to make:

“... all establishments falling within the scope of IDR Act are required to file a monthly production return to the department of industrial development. For the 290 items that are included in the latest IIP series, the reporting units are re­ported to be in the region of 10,000. However, since there is hardly any evidence of prosecution having taken place for the defaulting units, the supply of information by the establishments even in respect of the items that are includ­ed in the Index may not have been complete or timely. Since elimination of licensing for the most products and winding up of the erstwhile DGTD, the quality of data may even further deteriorate, despite the fact that the format of the return has been substantially simplified from its earlier complicated version (which in itself must have been a major contributor to the poor quality of data in the past). [Office of the Economic Adviser, Ministry of Industry 1995.]”

Another paper prepared by the officials of the CSO and presented in the same seminar, showed further deteriorating scenario in the quality of industrial data after the liberalisation:

“After the liberalisation, some of the major units including some of the PSUs have not been furnishing returns. This makes the estimation procedure for non-responding units very difficult in the absence of information on whether the unit is in existence or closed or on strike or on partial opera­tion. The government’s efforts to persuade the units to furnish returns now meet with little success. The GoI have subse­quently issued a Press Note No 8/92 dated 20.5.92, reiterat­ing the requirement of submission of returns by the indus­trial undertakings to the concerned technical agencies. Despite this, the coverage of units has been steadily de­clining [Kulshreshtha and Kolli 1995]”

It appears that the situation has become still worse in respect of the newer units. After the liberalization, the procedure of issuing registrations and making them regular after commencement of production has been done away with. This would mean that units could do away now even without informing the Ministry of industry about the progress report or the actual commencement of com­mercial production. It appears that there is very little data flow from new units after liberalisation (ibid).

            Report of the National Statistical Commission (2001) has pointed out the deficiencies in the existing system and recommended certain remedies. According to the Commission, deficiencies observed were absence of a comprehensive list of factories, inadequate response from data sources, reporting of abnormally high or low production figure, limitation of sample size – five units for about 45 items and only one unit in case of three items-, inadequate representation of both organized and unorganized sectors and weaknesses in the data collection mechanism. In order to rectify identified deficiencies, the Commission has given the following recommendations

Ø      The source agencies should expand their database to capture new units and items. To facilitate this, the CSO should regularly provide the list of items that are just below the cut-off criteria of item selection and may appear in the revised item basket

Ø      The source agencies should also identify important and fast moving items to include in their database for administrative purpose and revision of IIP.

Ø      The source agencies should preferably avoid the inclusion of items for which a very small sample is available but if it is necessary to include such items, the source agencies should closely monitor and collect data.

Ø      In order to solve the problem of non-response, the technical experts from the fields of industry and statistics should explore statistical methods.

Ø      The base year of the index should be revised quinquennially by the CSO to accommodate the structural changes that take place in the industrial sector.

Further, the Working Group on Commerce, Industry and the Corporate Sector constituted in the Modernisation of Statistical System in India under the Chairmanship of Dr Arun Ghosh has recommended for an alternative IIP based on large & medium industries and a separate quarterly index for Small-Scale Industries on experimental basis for one year or until the revision of IIP.

[An original note on the subject was published in the Economic and Political Weekly (EPW) of May 23, 1993. The same has been thoroughly revised and updated by Dr L Thulasamma for publication in this website]

Notes

1 The mining items excluded are moulding sand, emerald (dressed), cordierete, corundum (ruby) and staurolite for which production figures are not found on a regular basis.

2 subsequent to the south Asian financial crisis, the IMF introduced special macroeconomic data reporting system for its member countries, called special data dissemination standards (SDDS). India is a signatory to the SDDS. For details, see EPWRF (2003).

3 The original RBI study on the subject argued thus: “The index numbers are published according to the Standard Industrial Classification of industrial production items. For analytical purposes, this classificatory system is not entirely adequate. In view of the increased coverage of items, it is now possible to evolve alternative classifications, which would be useful for detailed analysis of trends in industrial production. Accordingly, it is proposed to set up some use-based, input-based and sectoral indices by rearranging the items, sub-groups or groups and aggregating these to meet the requirements of economic analysis. It may be mentioned that the Reserve Bank of India is already publishing in its Bulletin use-based indices classified into four major groups, viz, Basic, Capital goods, Intermediate goods and Consumer goods industries based on the revised index (1960 series). In addition, it is now proposed to have a sub-group of Consumer durables under the last major group.” (RBI Bulletin, June 1989: 989-1001)

4 In respect of the industrial products reserved for the small-scale sector, there is a major flaw in data arising from the fact that medium- and large-scale units reportedly undertaking produc­tion activities on the sly cannot include their unauthorised output figures in their monthly reports submitted to the govern­ment; incidence of such underreporting is said to be very sig­nificant. EPW Research Foundation (1996) presented tabular data in respect of the 18 SSI items and brought out how the medium and large units in the reserved areas were violating the reservation norms.

REFERENCES

EPW Research Foundation (1996): ‘Economic Reform and Rate of Saving’, EPW, Special Number, September.

 - (2003): ‘International Reserves, Foreign Currency Liquidity and External Debt: Data Template’, EPW, Special Statistics - 35, September.

Kulshrestha, A C and Ramesh Kolli (1995): ‘Impact of Liberalisation on Data Collection’, paper submitted at the Seminar of the Indian Association in Research in National Income and Wealth held at the ISI, New Delhi, March 24-25.

Industrial Statistics Division (2001): ‘An overview of Revision of Base Year of All-India Index of Industrial Production’, Central Statistical Organisation. 

Office of the Economic Advisor (Ministry of Industry) (1995): ‘Industrial Production Statistics’, paper submitted at the Seminar of the Indian Association in Research in National Income and Wealth held at the ISI, New Delhi, March 24-25.

RBI (1970): ‘Index Numbers of Industrial Production according to Alternative Classificatory Systems: A Methodological Note’, Reserve Bank of India Bulletin, June 989-1001.