Opinion | Politics should not meddle with our official statistics4 min read . Updated: 19 Nov 2019, 11:18 PM IST
Indian surveys were once hailed for their high quality but their reputation has taken a hit over the years
The National Statistical Office is making headlines again; this time, over the non-release of the results of the 75th round (2017-2018) of the National Sample Survey (NSS). Its report allegedly revealed a decline in average monthly per capita consumer expenditure (MPCE) in real terms compared to 2011-12. Faced with criticism, the government decided not to release the results, and indicated that it is examining the feasibility of conducting the next survey in 2020-2021 and 2021-22.
The latest controversy again puts a spotlight on what T.N. Srinivasan and others refer to as India’s downfall “from being the world leader in surveys" to a country “with a serious data problem". Indeed, as S.L. Shetty pointed out, India’s “official statistical collection machinery has been in decline for more than two decades". Our governments have repeatedly interfered with various official statistics. The NSSs have been in the limelight because of this for almost a decade.
Poverty reduction has been integral to the self-image of independent India. The NSSs are the primary means to track household consumer expenditure and poverty. Led by P.C. Mahalanobis initially, several economists and statisticians contributed to the design and development of the NSSs. American statistician Harold Hotelling is said to have remarked, “No technique of random sample has, so far as I can find, been developed in the United States or elsewhere, which can compare in accuracy with that described by professor Mahalanobis." However, beginning with the 1970s, political interference began to corrode trust in government statistics.
The slogan “Garibi Hatao!" played an important role in Indira Gandhi’s landslide victory in the 1971 general election. But her government soon came under pressure due to rising inflation and unemployment. In 1973, in the run-up to the Fifth Five Year Plan (1974–79), B.S. Minhas resigned from the Planning Commission over differences on the misuse of data to present a rosy picture of the economy. This was when the National Sample Survey Organization carried out two large-scale surveys in quick succession. The 27th round of 1972-73 was followed by the 28th in 1973-74, possibly because 1972-73 was a drought year. The Task Force on Minimum Needs and Effective Consumption Demand relied on the 28th round, even though it might not have been able to account for seasonality.
Two decades later, when governments were struggling with lacklustre outcomes of economic reforms, the 55th round (1999-2000) of the NSS stirred a controversy over the lack of inter-temporal comparability of its MPCE estimates. In this context, Angus Deaton and Valerie Kozel pointed out that “the political right had an interest in showing low poverty, and the political left in showing high poverty, and this undoubtedly intensified the debate on survey design and led to the unfortunate compromise design that temporarily undermined the poverty monitoring system".
Those contestations played out again a decade later. The 66th round of the NSS (2009-10) proved controversial because it showed that employment generation fell significantly short of the target of the 11th Five Year Plan. The then deputy chairman of the Planning Commission argued that the results of the survey “were not the best judge of the extent of impact [of] government policies… because it was a drought year". Official statisticians argued that since the GDP series could not be rebased using data from an “abnormal year", another large scale survey was conducted in 2011-12. The government also delayed the fourth round of the National Family Health Survey, which was eventually held after the 2014 polls. A 2011 Economic & Political Weekly editorial pointed out that “A pattern seems to have emerged of the government wanting to dismiss its own data... Doubts were cast earlier on the poverty numbers, then on the inflation indices and now it is the turn of the employment...."
The Bharatiya Janata Party jumped on the bandwagon and attacked the Congress. Perhaps it hadn’t expected that the boot would soon be on the other foot. Ahead of the 2019 polls, the Narendra Modi government delayed the release of the results of the 2017 Periodic Labour Force Survey that contested its employment claims. Two members of the National Statistical Commission resigned in protest. Now, unlike the 27th and 66th round when the results were released but superseded by fresh surveys, the Centre has gone a step further by deciding not to release the results of the 75th round of NSS.
The NSS is not the only casualty of this government’s cavalier approach to official statistics. Several other surveys and committees have seen their reports either delayed or trashed, while revisions have eroded trust in national accounts. The government also delayed the release of several tables of the 2011 Census, which should have been made public by Manmohan Singh’s government.
Unless statistical institutions are insulated from political interference, the government of the day, irrespective of its colour, would find it convenient to interfere. Incumbents tend to interfere towards the end of their tenure to burnish their credentials, while those who come to power promising the moon interfere after elections.
The government’s contention that the NSSs are unable to capture changing patterns of consumption is not entirely untrue, but that does not obviate the need to shield statistical bodies from politics.
Ankush Agrawal and Vikas Kumar teach economics at IIT Delhi and Azim Premji University, respectively