How reliable are India’s official statistics?

Authors: Ankush Agrawal, IEG, and Vikas Kumar, Azim Premji University

Developing nations like India are data intensive: they need socio-economic information about their population to design redistributive policies.

And, given their obsession with global rankings, they also need information to compare themselves with other countries. But India’s official statistics are not free of errors.

In July 2011, the governor of the Reserve Bank of India expressed concern over the quality of statistics collected by government agencies. A few months later, the commerce secretary admitted that India’s export figures for the April–October period were inflated by US$9.4 billion due to a misclassification of certain items and data-entry errors. Not long afterward, the chief statistician conceded that the accuracy of the Index of Industrial Production is questionable. And now the Planning Commission’s deputy chairman is arguing that the National Sample Surveys systematically underestimate household consumption. Concerns have also been raised about the duplication of efforts to collect statistics across various government departments, the inaccessibility of national data archives, and the infringement of privacy by the government’s data-collection machinery. To address some of these problems the Ministry of Statistics and Programme Implementation released a new data policy in 2009 and other related rules in 2011. More recently, the government also passed the National Data Sharing and Accessibility Policy in early 2012.

But problems with India’s official statistics run deeper. Take the example of the northeast Indian state of Nagaland. According to census data (which is collected every 10 years), in the decades 1981–91 and 1991–2001, the state’s population grew by 56.09 and 64.53 per cent respectively. But the 2011 census reveals that between 2001 and 2011 Nagaland’s population shrunk by 0.47 per cent. According to these figures, Nagaland’s population has declined in absolute terms in the absence of war, famine, natural calamities, political disturbance and dramatic changes in the socio-economic correlates of fertility. This is unprecedented in the history of independent India.

Demographic factors like migration or birth and death rates cannot explain why Nagaland’s population growth rate increased steadily until 2001, or why there has been a sudden drop in growth since then. Nagaland has not yet reached a level of socio-economic development that would warrant transition to a low birth-and-death rates regime. This calls for an examination of earlier censuses to determine whether Nagaland’s population was overestimated. Our research into this peculiarity collates data from various demographic, electoral and educational sources dating back to the 1960s in order to compare the estimated size of different sub-groups in Nagaland’s population.

We found that population estimates in censuses conducted between 1981 and 2001 lack internal consistency and contradict estimates from other data sources. In fact, during this period the census information for Nagaland defied both the projections of expert committees and the trends in socio-economic correlates of fertility, which suggested a lower growth rate. There exist pervasive inconsistencies in the population estimates of different sub-groups of Nagaland, which cannot be explained by demographic factors. Examination of other explanations for these inconsistencies suggests that the political geographic hypothesis — that people migrate to cope with arbitrary post-colonial boundaries — cannot explain the abnormal changes in Nagaland’s population if only intra-national boundaries are considered. While there is no reliable data to examine the international aspect of this problem, the available evidence does not support the political geographic hypothesis.

Nagaland’s chief minister, Neiphiu Rio, proposed a political economic explanation, suggesting that competing sub-groups in Nagaland’s population inflated their numbers to seek greater political representation and, by implication, a greater share of the state’s resources. According to Rio, the impending delimitation of constituencies in 2002 prompted Nagaland’s tribes to inflate their numbers in the 2001 census to avoid losing political representation to competing tribes and non-tribal plainsmen. But inter-tribal conflict and litigation forced the delimitation in Nagaland to be deferred until after 2031. Consequently, there was no incentive to inflate population figures in the 2011 census.

We found that inflated population figures in the 2001 census, evident across Nagaland’s three broad geographic and ethnic divisions, correlated with the expected loss of political representation due to impending delimitation. Moreover, the deflation in population figures for the 2001–11 period is related to the inflated data in the preceding decade. So, political economic factors can partially explain the anomaly in the 2001 census.

Scenarios of population growth developed on the basis of demographic, political geographic and political economic explanations reveal further inaccuracies. Contrary to the widely held belief that only the 2001 census is flawed, Nagaland’s population was also overestimated in 1981 and 1991. In fact, even the population figures for 2011 seem to have been slightly overestimated.

Long after the state government rejected the figures from the 2001 census though, government and non-government organisations have continued to use Nagaland’s flawed population series for sampling and analysis. The inconsistencies in successive censuses — the most important source of information about smaller Indian states such as Nagaland — and the government’s uncritical use of dubious statistics raise questions about the Indian government’s institutional capacity to design empirically informed policies.

Ankush Agrawal is Assistant Professor of Economics at the Institute of Economic Growth, Delhi.

Vikas Kumar is Assistant Professor of Economics at Azim Premji University, Bangalore.