Recent accusations of data manipulation do no good either to the Modi Government or the country. The refusal to face the truth will hurt the economy
Last week, over 200 researchers, academicians and planners from India and abroad signed a petition, demanding that the Union Government release the survey on household consumption expenditure conducted by the National Sample Survey Office (NSSO), now the National Statistical Office (NSO), which has allegedly been withheld by the Ministry of Statistics and Programme Implementation (MoSPI). Academicians have also urged the Government to make all other survey data public. A few days ago, the MoSPI had announced its decision to not release the NSO’s household consumption expenditure data for 2017-2018. According to a report that was leaked to the Press, the monthly per capita consumption expenditure by an Indian fell by 3.7 per cent. The NSO report was supposed to be released in June this year but was scrapped on account of data quality issues. According to the Government, there are sharp differences between estimates of the NSO survey data and that of the National Accounts Statistics (NAS).
Any data user in the country knows how the Indian statistical system works. In fact, consumption surveys are known to give results that diverge from macro-economic estimates of the national accounts. While the NAS presents the estimates of macro-economic aggregates, the consumption expenditure is calculated on micro-level data, based on the household survey. Therefore, the Government’s reason that it has withheld the reports because of “data quality” issues does not hold ground.
With the Government making tall claims of its successful performance on the economic front, the decline in average consumption must have come as a real embarrassment for the simple reason that this happened for the first time in the last four decades and supports the argument that the economy is indeed seeing a downturn. According to the last NSSO report in 2011-12, the per capita monthly consumer expenditure stood at Rs 1,501 and the leaked report has pegged it at Rs 1,446 for 2017-18. This shows a sharp decline in consumption expenditure in the period during the Modi 1.0 regime. On the economic front, contrary to the Government’s claim of a 6.5 per cent growth rate, the actual figures are off the mark. Since demonetisation and the Goods and Services Tax (GST) were the two major headwinds that held back India’s economic growth, it would have been more embarrassing for the Government to admit a reduction in household expenditure after these two major economic steps of the Modi regime backfired.
The NSO collects data on various aspects of the economy from time to time by taking periodic sample surveys. While small sample surveys are annual, large surveys, like on consumer expenditure, are conducted every five years. The NSO’s most vital survey is on household consumer expenditure, which forms the basis for assessing the country’s economic progress, the success or failure of poverty alleviation schemes, estimates of Gross Domestic Product (GDP) and disparity between various economic/social groups, among other things. To recall, the NSSO’s household consumer expenditure survey for 2009-10 had become a subject of much controversy amid questions over poverty estimates, which were based on the survey.
The then UPA-II Government had appointed a committee headed by economist C Rangarajan to review the methods used for calculating poverty ratio. The 2009-10 survey was cancelled and a fresh one ordered. It was only in January 2015 that a new series of NAS was introduced. It is not yet clear whether the gap between the 2017-18 NSSO data and NAS was due to this change. The NAS’s new survey was based on new base-year and the method of estimating the GDP, too, was revised.
No wonder then, the 2017-18 report was delayed. After 2011-12, the next round of data collection should have taken place in 2016-17. But now, while the 2017-18 data has been scrapped, the Government is exploring the feasibility of conducting the next Consumer Expenditure Survey in 2020-2021 and 2021-22 after incorporating all data quality refinements in the survey process. This means that new data will not be available before July 2023. This is a travesty considering that India is one of those countries where there is no official data collection of household or individual income. And the entire economic planning is done on the basis of consumption expenditure data. This reflects poorly on a country that once boasted a transparent system where the methodology and estimates of the surveys were clearly understood and were applauded by the international community, too.
The is not for the first time that the NDA Government has undertaken this data manipulation exercise. There are many incidents in the recent past where the data produced by the Modi Government have been a subject to controversy. Questions were raised on former Chief Economic Advisor Arvind Subramanian’s dramatic claim that the real GDP grew at a rate between 3.5 per cent and 5.5 per cent during 2011-12 to 2016-17. After quitting his post, Subramanian had questioned the methodology used to calculate GDP. Using other macro-level indicators, he showed that actual growth rate could be much lower than claimed. Since this data was based on new series of NAS, many pointed fingers. It should also be mentioned that the back series calculations done by the Central Statistical Office (CSO) were different from the one presented by the National Statistical Commission (NSC). This issue was also much debated among economists working on Indian data. Even data collection for GDP estimates has not been free from blemishes. In 2016-17, the NSSO found that more than 35 per cent of the firms, from where the data was sourced, were either non-existent or wrongly classified in the database.
Controversy also arose over the alleged suppression of the findings of the annual Periodic Labour Force Survey (PLFS) on labour force. Divergent views were expressed by the Government and independent researchers on the rate of unemployment among all social groups. In this case, too, the Government tried to suppress the data, which was already made public, unofficially. Ultimately, the data collected by NSC was held up and the Government was accused of manipulating data.
Over the years, there have been improvements in the timeliness of data compilation and dissemination. But the Indian statistical system is yet to match global standards. Most macro reports on all aspects of the economy are regularly published every year. This may be due to statutory requirement of placing annual reports before the Parliament. The most crucial weakness pertains to the survey based on data and the periodicity of its release. Most surveys, where data has to be gathered from the public, is conducted every four or five years. Its publication takes about 18 months (on an average). There is sufficient scope for improvements in the way we collect and disseminate data.
The NSC, too, remains mired in controversies. After its chairman and an external member resigned, expressing disappointment over the treatment being meted out by the Government, questions have been raised on the internal functioning of the organisation. It is welcome that the Government gave more teeth to it with a legislation. In order to streamline the functioning of data collection, the two arms of the MoSPI — NSSO and the Central Statistics Office (CSO) — were merged a few months ago. This move will likely foster better coordination as the NSO now comes directly under the Government, headed by the secretary of MoSPI.
It is unfortunate that the NSSO data on household expenditure has been withdrawn altogether. Releasing data selectively, which ensures positive results, and at the same time withholding uncomfortable data reflects the lack of honesty on the Government’s part. It also puts a question mark on the credibility of official statistics, which will always be a matter of suspicion now. Statisticians and economists know the limitations of data based on sample surveys. MoSPI should have released the data on time with a disclaimer about the limitations. Data delayed is data denied. The Government must deal with real problems head-on and find tangible solutions instead of hiding behind the bush.
(The writer is a professor of statistics and marketing research, International Management Institute, New Delhi)