When data becomes political
In the era of fake news, what constitutes the truth is based not just on facts but also where it is coming from. Governments have a tendency to project facts which suit their purpose but this is largely done by presenting that side of the truth which fetches them political mileage. While there has rarely been any attempt to manipulate statistical data, deliberately misleading data being presented to suit political purposes is not uncommon. The recent attempt to present employment estimates based on half-baked data is one such attempt. However, the statistical system has survived, primarily because the system of statistical data collection while working closely with the government has also managed to retain its independence. The credibility of the Indian statistical system was respected the world over for the quality of data collected. And also for the truthfulness of the data reported and presented.
But this has started to change. While the statistical system remains robust and has managed to retain its edge despite pressures from the political system, it has come under attack from various quarters. While only some of it is genuine criticism, matters have become muddled because of the reluctance on the part of the government to stand by the truth, whatever it is. Unfortunately, this trend has become prominent since the late 1990s with successive governments only going a step further from previous ones.
Five of the most important indicators of any economy and more so India would be, without much debate, gross domestic product (GDP), employment, poverty, health and education. One may add more to this list but these five represent the five most important aspects of the economy covering growth, jobs, poverty and human development. On each of these there have been attempts to discredit, delay or disrupt data collection and dissemination.
Statistics of growth as measured by national accounts are the most used headline numbers. Despite flaws, which are inevitable in a country with a large informal sector, there were few questions about the credibility of the data until 2012-13. Not so in recent times, with the data being questioned by the government’s own economic adviser, the Reserve Bank of India, independent researchers, and by international agencies such as International Monetary Fund (IMF). Such questions gain credence because the Central Statistics Office (CSO) has failed to convince users about the comparability of the new GDP series. The delay in bringing out the back series based on the new methodology, a prerequisite whenever a new methodology is brought in, only added to the doubts. It has been four years since the new methodology was adopted. The end result is that there cannot be any meaningful analysis of how the economy responded to different events—short-term as well as long-term.
So is the case with the data on jobs, which are now being debated academically as well as politically. While the government intervened to stop the annual survey by the labour bureau after it showed a sharp decline in employment by 16 million, it also delayed the usual employment-unemployment surveys. The end result is that this will be the first time that an elected government will seek re-election on an issue of public importance without any credible official data. Between 2004 and 2011-12, there were six surveys on employment and unemployment by the NSSO. There are none after 2011-12.
The case of poverty estimates is similar. The last estimates were based on the Tendulkar Committee report which were contested by civil society and the then opposition led by Bharatiya Janata Party (BJP). A new committee under C. Rangarajan submitted its report to the present government in 2014 but the government has ignored the recommendations. It has also not yet decided the methodology even though data from consumer expenditure will be available sometime later this year. Another example where the citizens will be clueless is what happened to poverty during this government’s tenure when the election is likely to be fought on the pro-poor credentials of the government.
On education, the only information on learning outcomes is available from the Annual Status of Education Report (ASER), conducted by a private body. The Indian government agreed to take part in the Programme for International Student Assessment (PISA) surveys in 2009. Even then the government granted permission for the survey in only Himachal Pradesh and Tamil Nadu, two of the best states in terms of educational outcomes. The results showed India at the bottom of the international ranking. The reaction of the then government was to pull out of the surveys.
On health, the only credible survey on malnutrition, health outcomes and women is the National Family Health Survey (NFHS) as part of the global Demographic and Health Surveys (DHS). The first one was held in 1992, the second in 1998 and the third in 2005. Normally, they are held every four-five years. The negative media coverage of malnutrition numbers from NFHS-3 meant that government decided to not hold the fourth one. It was only after pressure from academics and civil society that the government finally conducted the fourth survey in 2015-16. While there is finally some information, it also means that information on such a crucial variable which should have been available on a regular basis is only available after a gap of ten years.
So is the case with tax data which the government stopped publishing in 2000 when the results from the tax data analysis by Abhijit Banerjee and Thomas Piketty from a carefully maintained series since independence showed that inequality has gone up. Again after pressure, the government has finally started releasing data from 2012-13 but absence of data between 2000 and 2012 means that the analysis is limited by data considerations for a period when inequality rose sharply.
Same is the case with the Socio-Economic Caste Census which conducted for the first time a socioeconomic survey of households by caste. Even after five years of the data being collected, nobody knows what happened to the caste data. At a time when several caste groups are arguing for reservation, the data could have provided useful information. But there is no information on the data even though crores were spent.
At a time when the government is arguing for using “big data” for policymaking, its own actions suggest scant regard for a statistical system which has been built by stalwarts through decades of hard work. The reasons are not ignorance or apathy but a disdain for any truth which is uncomfortable. The independence of the statistical system and credibility of data on basic indicators of the economy is not just a need of the academic community, it is also fundamental to the functioning of democracy. Denial of access to credible data and manipulation of statistics not only weakens governance, it also denies the citizens the basic right to hold the government accountable, the essence of democracy.
Himanshu is an associate professor at Jawaharlal Nehru University and visiting fellow at Centre de Sciences Humaines, New Delhi.
- US FTC probing Facebook data scandal: report
- NMC Bill shouldn’t make bridge course for AYUSH practitioners mandatory: panel
- Jeff Bezos ducks tech doldrums as Larry Ellison’s wealth falls $4.5 billion
- Breaking up with Facebook will be harder than it appears
- Amazon tops Alphabet to become second most valuable US-listed company