When Evidence is Anti-national
The NDA government has recently declared that it had made an unannounced decision to trash the ‘draft’ report of the official consumer expenditure survey relating to 2018-19. These benchmark surveys, normally conducted every five years in the past, are the most reliable sources of information on consumption trends, that also provide the basis for estimates of the proportion of people below a chosen poverty line. The last such survey was conducted in 2011-12. The government’s announcement came in the wake of a leak of a copy of the report that had been finalised by relevant wings of the now restructured National Statistical Office (formerly National Sample Survey Organisation) under the Ministry of Statistics and Programme Implementation.
The leak had revealed that ‘real’ monthly per capita consumption expenditure (MPCE)—or nominal consumption adjusted for inflation to make figures comparable across time—had fallen by 3.7 per cent from Rs 1,501 in 2011-12 (when the last such survey had been undertaken) to Rs 1,446 in 2017-18. This absolute decline in real per capita consumption is an unusual occurrence reflective of significant economic stress.
Not surprisingly, the government’s decision to hold back the results of the survey has been widely interpreted as one more move on its part to supress information indicative of poor economic performance under its watch. This tendency to suppress officially generated evidence is indicative not just of the proclivity of the NDA government to rely on propaganda rather than good policy to garner popular support, but also of its desire to conceal any evidence that its two ostensibly “game-changing” policy interventions, demonetisation and the hastily pushed GST regime, have had damaging economic consequences.
That the NDA government believes that any evidence contradicting its hype about the economy does not deserve to be in the public domain, has been illustrated by two kinds of official interventions in the statistical realm in the recent past. One is to ‘revise’ officially generated data in order to change the picture it presents of the economy. This is what has happened with respect to the data on GDP, which the government cannot but release, given its decision to adhere to data dissemination standards set by institutions like the International Monetary Fund. Not satisfied with the rates of growth reflected in the GDP numbers, the government decided to opt for major revisions of the source data and methodology used in the estimation of GDP in the new series with 2011-12 as base year. The revision not only raised rates of growth of GDP in some years that overlapped with the older series, but placed India among the fastest growing economies in the world during the first Modi regime. In fact, growth during the demonetisation year 2016-17 was also shown as being as high as 7.7 per cent, despite evidence of demonetisation-induced contraction in a wide range of sectors. Since other numbers did not seem to support this evidence of stellar GDP performance after 2011-12, doubts have been expressed in a wide circle of the credibility of the new series. Even Arvind Subramanian, the Chief Economic Advisor during the first Modi regime has expressed such scepticism, and put out a paper saying actual GDP growth in India was around a half of what the official figures were claiming it was.
The government’s response was that the new GDP series had been computed using international best practices. But that claim was undermined by its own actions. The change in methodology had made it difficult to generate a comparable “back series” of GDP figures for the years prior to 2011-12. So, an official committee was set up to work out a methodology to estimate those comparable historical numbers. When it prepared its report, the estimates generated showed that growth performance was better under the UPA government than under the first Modi government. Despite the committee being official, the government decided to ignore its report and then reject it. It relied instead on a hastily evolved alternative method to generate a back series that painted a better picture of performance under the first NDA government and reversed the finding that relative performance was better under UPA. What became clear was this alternative back series was not the result of the work of the statistical agencies alone, but was prepared with the involvement of a politically aligned user organisation like the Niti Ayog.
Revising GDP numbers, even if controversial, may be easy. But results from large scale surveys conducted by the National Sample Survey Organisation and its less independent successor, the National Statistical Office, which adhere to processes and vetting mechanisms honed over the years, are more difficult. So when the employment and unemployment survey (now rechristened as the periodic labour force survey) relating to 2017-18, yielded results that indicated that the unemployment rate in that year was at its highest since such surveys had begun to be conducted in 1972-73, the government was faced with a challenge. Revising those results would require tampering with the schedules, which was not feasible. On the other hand, with an election impending, it did not want an official report to contradict its claims of having fulfilled its employment generation promises. So, it decided to hold back the survey. Unfortunately for it, Somesh Jha of the Business Standard managed to access a copy of the report and put out the main results, forcing the government to state that what Jha had obtained was only a draft report which was not yet official, even though the National Statistical Commission had cleared the report for release. The government even suggested that data quality issues warranted re-examination of the results. However, once the elections were done and the results announced, the report was released in the form in which Jha had accessed it, making clear that the report was held back on purely political considerations.
This history makes clear that the tendency to identify those who differ with the government reading of any development—political or economic—as anti-national, and dissenters as ‘urban naxals’, has now spilt over into identifying any evidence that counters the government’s hype as wrong and/or not worthy of public scrutiny. It is that tendency that resulted in the holding back of the consumption survey relating to 2017-18, which suggests that poor economic performance and the effects of demonetisation and the GST had resulted in distress of an intensity that had resulted in a drop in average per person consumption spending in 2017-18 relative to 2011-12. But, once again, Somesh Jha has managed to access a copy of the report, and make its key results available over four articles in the Business Standard.
This time, however, the government’s response has been more aggressive. An official statement from the Ministry of Statistics and Programme Implementation in response to Jha’s reports holds that “the results of the survey were examined and it was noted that there was a significant increase in the divergence in not only the levels in the consumption pattern but also the direction of the change when compared to the other administrative data sources like the actual production of goods and services. Concerns were also raised about the ability/sensitivity of the survey instrument to capture consumption of social services by households especially on health and education.” It claims that the matter had, therefore, been referred to a Committee of experts which noted the discrepancies and came out with several recommendations. It transpires, however, that the minutes of the meeting of the committee of experts suggest that they had not recommended trashing the survey and not releasing the figures. So the only reason for holding back the consumer expenditure survey results is that, since, on the grounds that 2017-18 is not a good year, the government has decided that that financial year would not be the base for a new revised series of national accounts statistics, the absence of a consumption survey for 2017-18 does not create a problem. That, however, leaves unexplained why the survey results could not have been released for use for other purposes.
In sum, the whitewash of the decision to supress the consumption data does not convince. Discrepancies in aggregate consumption figures generated for the National Accounts, using a different estimation methodology, and that yielded by the consumer expenditure surveys is an old problem and is not confined to India. And consumer expenditure surveys are not meant only as the basis for national accounts estimations, but serve other purposes including estimation of changes in the level of poverty in the country. Initial calculations suggest that even though the reduction in consumption has occurred even in the middle and lower-middle income groups, reflecting the widespread adverse impact of demonetisation, the fall in per capita consumption has affected the lower income groups enough to reverse the long term decline in the incidence of poverty in India. The government is naïve if it believes that treating the release of “inconvenient” data as ant-national and supressing it would absolve it of being responsible for this crisis situation.
(This article was originally published in Asiavillanews on November 21, 2019)