An important issue the new central government will face is the need to revise the base year for the purpose of Gross Domestic Product (GDP) and other key economic calculations. A base year is the reference year whose prices are used to calculate, importantly, the real GDP growth rate. Such base year decisions are also relevant for other metrics such as the Index of Industrial Production, the Wholesale Price Index and the Consumer Price Index.
India currently follows the base year of 2011-12. The Central Statistical Office (CSO) in January 2015 announced 2011-12 as the base year for India’s national accounts. The revision replaced the previous base year of 2004-05 and was in line with the recommendations of the National Statistical Commission, which advocated base year changes every five years.
Choosing the base year and revising it are important for any economy, since they track structural changes. Imagine, as an extreme case, that the base year of India had been unchanged at 1980-81 (an earlier series, which was later replaced by the 1993-94 series). The resultant GDP data with such an old base year would have been unable to capture the huge structural transformation that India underwent, especially since 1991, with the growth of the IT sector. A revision of the GDP and updating the base year thus allow statisticians to reweight the relative importance of different sectors of economic activity. It also allows for further change or reconsideration of methods and data sources.
Base year revisions are very different from annual revisions in the National Accounts. An annual revision merely incorporates changes in the data available and updates such data, but does not tamper with the basic conceptual framework. In order to ensure a strict comparison over years, it also does not use any new and better data sources. Base year revisions, on the other hand, not only shift the reference years, but also use updated surveys and studies, and follow international guidelines to incorporate conceptual changes.
India, since 1948-49, had been undertaking base year revisions once every 10 years. Such a long gap was justified since the structure of the Indian economy underwent very little change over the next three decades. The base year for the economic metrics was aligned with the year in which the population census was conducted. This was because information on the workforce had an important role to play in base year estimates and workforce estimates were obtained from the decennial census.
However, the CSO in February 1999 moved away from this practice and replaced the 1980-81 series with a 1993-94 one, instead of a 1990-91 series. This was done since the CSO replaced the population census with the quinquennial survey conducted by the National Sample Survey Organisation on employment. The latter was seen as providing better estimates of the workforce, especially the female workforce in rural areas. With this change, India started undertaking base year changes every five years from 1993-94.
While base year revisions followed in 1999-2000 and 2004-05, the next revision—due in 2009-10—was pushed to 2011-12. This was due to the global financial crisis in 2008, which would have affected the estimates in the following years as 2008 was not a ‘normal’ economic year. Similar considerations prevented a base year change in 2016-17, another abnormal year because of demonetisation and the resultant slump in economic activity. A host of domestic and global economic and non-economic considerations have led to India not making changes in its base year since then.
In 2019, the Indian government was reported to be considering a move to a chain-based mechanism—one that would allow GDP to be compared to the previous year, instead of having a fixed base that would need to be revised every five years. Advanced countries such as the United States, Australia, Canada and most of the European Union countries use the chain-based method.
However, such a move to a chain-based mechanism may be ill-advised for a country like India. As noted by the System of National Accounts (SNA, para 15.43) 2008, “If individual prices and quantities fluctuate so that the relative price and quantity changes occurring in earlier periods are reversed in later periods, chaining will produce worse results than a simple index.” In India, with its highly volatile and monsoon-dependent agricultural sector and its vulnerability to fuel prices, chain linking could lead to distortions when prices or volumes fluctuate.
There are other problems with using the chain-based method in the case of resource-starved countries like India. Chain-linked indices are stated to be computationally difficult, demanding additional resources. The 2008 SNA thus noted that annual chain indices require far greater computing compared to fixed-weighted indices, and should not be attempted without adequate, tailored software.
Besides, such chain indices do not allow for extrapolations back or forward from current values. Such extrapolations are important in order to have comparable data over time.
Thus, while a base year revision is warranted, given that India has not revised it for almost 10 years, we should desist from quick fixes based on advanced-country experiences. We should chart our own course. A fixed base year, revised every five years, still remains India’s best strategy.
(Views are personal)
(tulsi.jayakumar@spjimr.org)
Tulsi Jayakumar | Professor, finance and economics, and Executive Director, Centre for Family Business and Entrepreneurship at Bhavan’s SPJIMR