Opinion Amid debate on India’s statistical system, national data beyond surveys

First, we should recognise that there is a problem in the statistical system that needs to be fixed. Defending the statistical system is no solution at all. The National Statistical Office (NSO) has been collecting data primarily through administrative and sample surveys, both of which have their own strengths and challenges. The data collection from administrative sources is economical and less time-consuming, but has several challenges in terms of representativeness. Sample surveys are costlier, both in terms of money and time. The updation of the Census frame used for most surveys needs to be digitised dynamically and made accessible to improve the quality of surveys and reduce bias in the estimates. Geospatial technologies and crowd-sourced data platforms now permit such dynamic updation.

Second, the national statistical system needs to expand and diversify its resource base of data — it should include new and emerging sources like Big Data leverage processing through machine learning and artificial intelligence. The efforts by the NSO for developing “standards” and “methodologies” for data validation of these new datasets are going to be extremely important, so they supplement conventional data sources. The UN Statistics Division has come up with guidelines for using Big Data for official purposes. The NSO needs to work closely with multilateral and regional agencies for enhancing the capacity of the statistical system for the use of such data available from alternative sources.

Third, the strength of the national system is integrally dependent on the strength of the state statistical systems. In this direction, the Dholakia Committee Report 2020 on sub-national accounts is crucial — it could pave the way for state governments in pursuing and adopting a bottom-up approach, thereby strengthening the data collection capacities of the state governments. Several states are yet to initiate building institutional frameworks at the state and district levels. The Ministry of Statistics and Programme Implementation (MOSPI) launched the India Statistical Strengthening Project with financial support from the World Bank for enhancing the capacities of state statistical systems for data collection.

However, parallel efforts are needed to enhance and institutionalise inter-agency coordination covering both national and sub-national statistical systems. The ongoing debate also encompasses the rural-urban divide-related definitional issues. Without taking any sides, it is equally pertinent to mention that the rural and urban sub-groups appearing, howsoever mutually exclusive, are surfeit with enormous synergies necessitating a holistic approach to capture the interaction terms. Madhya Pradesh has taken the lead by establishing a permanent state statistical commission for improving and integrating the statistical data flow systems. Development of district domestic product using GSTN and revenue data as well as a state statistical business register has already commenced. Once established, this could pave the way for improving the statistical system in other states as well.

Fourth, some lessons may need to be drawn from our national experience of improving weather predictions. A few decades ago, the weather forecast used to be the subject of various jokes. The Ministry of Earth Sciences established the National Centre for Medium Range Weather Forecasting in 1988 and used India’s first supercomputer to develop and evolve advanced numerical models for weather forecasting. In the face of jokes and criticism, the commitment to the upgradation of observation systems has made the biggest contribution in enhancing predictability, along with an improvement in the capacity of human resources for complex data collection and the development of IT infrastructure. There are several other instances, both in the public and private sector, like in direct taxes, GSTN, railway reservations, banking and financial sector, UPI, portal aggregators, online shopping etc where technological advances are leveraged to improve service delivery.

Five, the underlying tenet of any official statistical system/product are the resources (physical, human, financial and technology) available to the system. A statistician learns very early that the efficiency, quality and timeliness of the product are dependent on the resources available and all efforts are made to optimise the same. We are well aware that economic growth occurs when there is a sustained expansion in the production possibility frontier and this happens when we develop better technologies; improve the quality of labour through education, on-the-job training, work experience and exposure to best practices; and acquire more machines/technology (capital) to help produce more. In the same analogy, there needs to be sustained growth in the resources available to the national statistical system for it to improve and this needs to be seen as an investment to ensure that India achieves the target of becoming a $5 trillion economy.

To catalyse and synergise these efforts, the National Policy on Official Statistics, announced in the Budget 2020 needs to be finalised quickly along with appropriate institutional support and resources. This will ensure that we are able to track India’s progress on the Sustainable Development Goals using a bottom-up approach and also ensure that no one is left behind. We are confident that fast-tracking reforms and investment in the national statistical system in a mission mode is the need of the hour and cannot be delayed if India wants to once again play an active role in the international statistical fraternity.

Chaturvedi is DG, RIS and Srivastava is Chairman, Madhya Pradesh Statistical Commission and former Chief Statistician of India and Former Secretary, Ministry of Statistics and Programme Implementation, MoSPI. Views are personal