Last week, data released by the Central Statistics Office (CSO) showed that factory output growth had slipped to a 17-month low in November. This growth is measured on the basis of the “Index of Industrial Production (IIP)”. What is this index, how is it calculated, and what do its readings tell us?
The index & its importance
The IIP is a composite indicator that measures changes in the volume of production of a basket of industrial products over a period of time, with respect to a chosen base period. It is compiled and published monthly by the CSO with a time lag of six weeks from the reference month.
The all-India IIP provides a single representative figure to measure the general level of industrial activity in the economy on a monthly basis. Used by government agencies including the Ministry of Finance, the Reserve Bank of India etc, for policy purposes, the all-India IIP forms a crucial input for compilation of Gross Value Added (GVA) of the manufacturing sector in the Gross Domestic Product (GDP) on a quarterly basis. It is also used extensively by financial intermediaries, policy analysts and private companies for various analytical purposes.
It is crucial considering the IIP is the only measure on the physical volume of production. While its impact on GDP calculations is lower following changes incorporated when the latest shift to the 2011-12 base year happened, it remains extremely relevant for the calculation of the quarterly and advance GDP estimates. For the annual revised estimates, in any case, the CSO used to replace the IIP with the ASI (Annual Survey of Industries), which comes out with a two-year lag.
Changes in base year
The change in the base year to 2011-12, which happened in 2017, was the ninth revision of base year of the all-India IIP since the beginning of its dissemination, with the previous ones being 1937, 1946, 1951, 1956, 1960, 1970, 1980-81, 1993-94 and 2004-05. While a change in the base year should not end up making too much difference to the IIP growth figures, the bigger impact is on account of the difference in the constituent items of the index and weights assigned to each of them.
In the 2011-12 series, as compared to 2004-05 series, a number of items were introduced or deleted which helped to check the volatility of the index that was earlier seen especially for capital goods. Items such as refined palm oil, cement clinkers and surgical accessories were introduced while tooth brushes, chewing tobacco, fans, calculators, pens and watches were removed. In all, 149 new items were added in the new IIP 2011-12 data series, while 124 of them were deleted. The items ‘salt’ and ‘coffee’ in the existing series have been replaced with ‘iodized salt’ and ‘instant coffee’ respectively, given their increasing importance in production. On the whole, the new series had 809 items from the manufacturing sector as against 620 from the old 2004-05 series.
Globally, the compilation of such indices dates back to at least the 1920s. A United Nations publication detailing the production index methodology, namely Index Numbers of Industrial Production, was published in 1950 — the first and only United Nations publication on this topic. While the UN has published material on related topics, such as the Guidelines on Principles of a System of Price and Quantity Statistics in 1977 or the Manual on Producers’ Price Indices for Industrial Goods in 1979, no revision of the methodology published in the original version of the index number manual has been released. Since its publication in 1950, many changes have taken place that require an updated version of the index number publication. This includes, on one hand, country experiences compiling index numbers over the past decades. Also, a number of underlying and related statistical standards and recommendations have changed, particularly in recent years, and the concepts and methodology applied in the original index number manual needed to be updated.
IIP & ASI
Since the ASI is the main source of long term industrial statistics while the IIP is a monthly indicator based on items and factories selected from ASI, comparisons between the growth rates of manufacturing sector based on the two datasets are obvious. The divergence in the data sets is primarily on account of the fact that the IIP is based on a fixed set of items and factories chosen in the base period whereas the ASI is a record-based survey of establishments registered under the Factories Act, 1948 in which the sampling frame and the sampled establishments undergo significant changes.
Consequently, the ASI captures information of new items and factories whereas the IIP does not. Also, the IIP is based on a much smaller sample of factories as compared to that of ASI. Growth rates in IIP are based on volume of production whereas growth rates in ASI are derived on the basis of Value Added (Output – Input). As these parameters are conceptually different, the resultant growth rates are also different. Further, establishments selected in IIP are generally larger in size whereas ASI establishments cover both large and smaller units. So, the growth rates in IIP are lower as the smaller units that have a thinner base and hence show higher growth.