Recently, there has been a spate of writing on India’s statistical system and the data produced by it. The ensuing debate has witnessed active participation from within and outside the government. While discussions on data and its veracity are welcome, the important point on the usefulness of available data to measure social and economic progress of a transforming economy looms large. The debate has also to be seen in the context of the increasing global prominence of the economy and India’s election to the highest statistical body of the UN for a four-year term.
Economic and social data ideally is a public good. Its provision, like many other public goods, is the responsibility of the government. While there are numerous private data-generating and publishing entities, they exclude data consumers through their pricing and restrictive access policies. Anticipating this, independent India planned for a well-designed statistical system and embarked on the journey of setting it up at both at the Centre and in the States. As noted by journalist Pramit Bhattacharya in a paper, “India’s statistical system was the envy of the world till the early 1970s”. However, over time, this saw a steady decline and has reached its present state, which is the subject of the current debate.
Three issues
Three points emerging from this debate warrant scrutiny. The first is the quality and credibility of data. Policy formulation, implementation, and research are not possible without accurate and credible data. Credibility is closely connected with the form of the data and the value of information, and is dependent on the reliability of the source. Accuracy is when the data is a true representation of the expected values. The issue of capturing data correctly is of paramount importance for surveys as they are the source for official rates of unemployment, poverty, and other statistics that guide policy. This data also aids economic research. Surveys, especially household surveys, are facing questions of credibility in terms of the samples selected, representation of geographical regions, urban and rural strata, and on the issue of non-response by households. Given the structural changes taking place in the economy, there can be an accurate portrayal of the economy and society only through a realistic sample. Without this, biases creep in. A recent paper by the EAC-PM argues that surveys do not adequately capture the urban part of the economy. While this is a crucial issue, the data and definitions of ‘urban’ itself need to be re-examined.
Second, there are multiple agencies providing data on the same set of indicators for the same sector, but giving different numbers. This poses the problem of choosing an appropriate data set to formulate policies. This is evident in the case of manufacturing. Data on this sector for calculating GDP is from the Ministry of Corporate Affairs’ MCA21 portal. But this data differs from the data published by the Annual Survey of Industries. Moreover, a 2019 report by the National Sample Survey Office on the services sector found that nearly 36% of companies in the MCA21 system that were used in computing GDP were either not traceable or not classified properly. As of June 2023, the MCA21 portal was plagued by technical glitches. There were problems in uploading filings and accessing documents. About 800-900 companies were being registered every day on the new portal of the MCA. However, there is concern about a large number of shell companies being formed, which undermines the use of this data for GDP calculations.
The competence and ability of the system to generate and disseminate high-quality data is the third issue. In recent times, not only have surveys been delayed, but the publication of processed data of a completed survey has also been withheld. This has hampered the generation of micro-level data on important variables such as consumption used for assessing the extent of poverty. This is also the case with macroeconomic data. The wholesale price series used to assess inflation and the consumer price index are overdue for revisions. To assess economic growth, national income estimates need revisions. Non-availability of updated data hinders the assessment of economic growth and poverty reduction.
Urgent attention required
The non-availability of credible, coherent data on the economy and society is a problem that needs urgent attention. With the 2021 Census yet to take place and with a re-look required for methods of surveys, data are not available for assessing economic performance and development outcomes. In the absence of such data, policymakers look for quick fixes and celebrate success on the outcomes derived from thin samples. Recently, NITI Aayog’s National Multidimensional Poverty Index 2023, which used data from the National Family Health Survey (NFHS) of 2019-21, showed reduction in multidimensional poverty since 2015-16. No doubt the NFHS has a robust methodology for collecting data for a specific purpose, but this particular index needs a caveat as it is based on a sample of 6 lakh households, while India has more than 30 crore households.
It is time for an overhaul of the system. We need better communication between the creators and users of data. There should be an attempt to move beyond dashboards and disseminate data more transparently. Going back to the Rangarajan Committee Report (2001) and taking stock of what has been done could be a good starting point.
