Opinion
India’s statistical system still has questions to answer
Summary
An institutional mechanism for independent audits can help restore trust in the country’s official statisticsI am glad to note that India’s longest-serving chief statistician, T.C.A. Anant, found some useful insights in my recent Carnegie Endowment working paper, India’s Statistical System: Past, Present, Future (shorturl.at/FIJOU). But he objects to the narrative of a long decline in India’s official statistical system in his Mint column last week (shorturl.at/chrM6). My paper suggests that India’s statistical system was the envy of the world till the early 1970s, and saw a gradual decline since then. Anant contends that the 1970s and 1980s saw several improvements in the statistical system. He cites the examples of the regional accounts committee in 1972 that helped standardize state domestic product estimates and the recommendations of the NSS review committee that led to the establishment of the National Sample Survey Organization (NSSO) in 1970.
Both these examples pertain to the early 1970s, and do not contradict the paper’s argument that the statistical system saw greater initiative till the early 1970s.
Second, Anant’s article seems to oversell the impact of the NSS review committee. The committee was supposed to help speed-up the publication of NSS results, as Anant correctly points out. But as the 2001 report of the Rangarajan commission noted, “the delays (in the publication of NSS results) persisted till the 90s, when with effective use of modern electronic computers the problem was finally resolved." The report argued that the “scientific approach of problem solving through analytical studies and pilot experiments, for which the NSS was well known in its early days, have been given up under the pressure of day-to-day work."
Third, Anant seems to have missed the central thrust of my arguments. The paper tries to assess whether the statistical system at a given point of time was able to respond to the current and anticipated needs of data users. It argued that the first two decades after independence saw many pioneering initiatives to address the data needs of economic planners. Technocrats and leading politicians of the day were attentive to the needs of the statistical system.
Once these conditions changed, the statistical system lost its dynamism. An official review of the statistical system in 1979-80 (Narain-Bhatnagar committee) expressed dismay over the growing gulf between the world of official statistics and that of statistical research. It highlighted the system’s inability to meet the data requirements of policymakers in a timely manner, and the lack of statistical inputs in decision-making. Unfortunately, key recommendations of that committee weren’t implemented. The apathy of the ruling establishment meant that the statistical system was ill-equipped to face the demands of a fast-growing market-driven economy in the 1990s. At that point, the statistical crisis was too severe to be ignored. The Rangarajan commission was set up to stem the tide.
That the statistical system showed greater initiative in the 1950s, 60s and early 70s is not a narrative spun out of thin air. It reflects the broad consensus among the respondents I have interviewed during my research, half of whom have spent their working lives in India’s statistical system. This consensus also mirrors the assessment of the Rangarajan commission.
This does not mean there wasn’t any attempt to reform the system or improve statistical products in the late 1970s and in the 80s. But the improvements were inadequate. The reforms initiated in the wake of the Rangarajan commission report have also been partial. Hence, the statistical system is still struggling to deliver the kind of high-quality and high-frequency datasets that citizens, policymakers and investors expect from it today.
Anant presents an incomplete view of the resistance to statistical reforms when he pins the blame on state governments. Several statistical reforms, such as institutionalizing independent audits of core statistical products, can be implemented without approval from state governments. The National Statistical Commission (NSC) was set up in 2005 with a mandate to conduct such audits. In 2011, it commissioned an exploratory audit of the Index of Industrial Production. It was to serve as a template for future audits to ensure data quality. It ended up being the last of its kind.
As I have documented in a detailed report earlier, the NSC was marginalized by successive chief statisticians (‘How India’s statistical system was crippled,’ Mint, 8 May 2019, shorturl.at/jkRV6). In its 2017-18 annual report, the NSC complained that it was not consulted by the ministry of statistics and programme implementation (Mospi) while announcing surveys/census. India’s apex statistical body was not even provided a functioning office, the report said.
Attempts to provide financial autonomy and legislative backing to the NSC failed to win support from the chief statistician, a former Mospi official M.V.S. Ranganadham has said on record. It is possible that successive chief statisticians did not see the NSC as a viable regulator. But it is not clear why they didn’t set up alternate audit mechanisms.
Finally, I must point out that my paper is an initial attempt to document the history of an under-studied, and sometimes under-appreciated, statistical system. I hope it will nudge others to examine India’s statistical governance in greater detail. Former Mospi officials can help ease the path of researchers by sharing greater details from their stints: the hurdles they failed to overcome, and the ones they successfully tackled.
Pramit Bhattacharya is a Chennai-based journalist. His Twitter handle is pramit_b
