The coronavirus pandemic poses a challenge for implementing developmental responses in the context of district, one of the most important administrative units in India. There has been a push towards development at the grassroots level through AtmaNirbhar Bharat, Vocal for Local, One District One Product and National Infrastructure Pipeline.
To measure the development and diversity of the country, it is important to focus on the statistical architecture of districts. Therefore, credible and assessable dynamic data disaggregated at the district level is vital. The need for this has been raised at various levels including policy. In light of this, a special lecture was organised on 'District development, data and diversity' followed by a panel discussion. The lecture was delivered by Prof. Abusaleh Shariff, US-India Policy Institute, Washington at a webinar organised by Impact and Policy Research Institute (IMPRI), New Delhi on November 2, 2020.
Development data in India
“There is a massive improvement in the size and quality of development data in India. Data is accrued from numerous levels in the country with a population of over 1.3 billion. With multiple sources, methods and the interplay of several levels, there are huge complexities involved in generating authentic data. While digitisation has helped to a huge extent, yet a proper evaluation of government data remains to be done,” says Prof Shariff.
Data aggregation, disaggregation and structuring are critical to be able to tell the truth through data. Getting data at the district level for policy analysis is not easy in India. Issues related to gender, diversity and demographic dividend are often ignored in data.
“It is very important to institutionalise independent data collection and warehousing outside the government system. This is primarily because data is a public resource, and must be easily available to all who seek to understand our society and economy,” says Prof Shariff.
Though we have national-level statistics, international comparisons, state comparisons, we fall short of data for the district and parliamentary constituency. There is a need to have data at this level too, as a lot of funds are transferred through the parliamentary system to the districts.
“Besides, multiple sources of government data such as the national sample surveys, we have other sources of data like private experts, data generators, data through academic research etc. But more qualitative and quantitative data is needed through rapid rural appraisals. We need multiple strategies to involve communities to understand the importance of creating data at the district level,” says Prof Shariff.
District development and data
“The attempt to identify “weak or backward” districts has continued post-1960s after the report of the committee on the dispersal of industries indicated how certain districts were affecting the country’s overall performance. There was a focus on specific districts state-wise, but the poorest districts were still left behind,” says Dr Avani Kapur, Director, Accountability Initiatives, Centre for Policy Research (CPR), New Delhi.
For successful policy interventions, especially in the social sector, robust planning and coordination in governance are needed at the district level, where most things converge. Often, the district remains just an implementing arm, without its resources and enough decision-making authority. This becomes a big problem when discussing inter-district variation.
“The Economic Survey of India, 2016 mapped the top six welfare programs: housing scheme, Sarva Shiksha Abhiyan, mid-day meal scheme, road scheme, MGNREGA, and Swachh Bharat Mission. Poorest districts did not receive even 40% of the total resources under any of these schemes. While in theory, we have the system of prioritisation at the district level, in reality, it does not happen. So, the poorest are often at the receiving end, which then becomes a continuous vicious circle,” says Kapur.
Avani Kapur further says that “a mere focus on district won’t work till budgeting and systems continue to be centralised. Systemic changes are required and the role of the district as a development function in a welfare state like India is not fully understood. Despite the 73rd and 74th Constitutional Amendments, we still haven’t placed enough emphasis on the capacity or role of the district administration.”
Most schemes do not converge even at the district level, while theoretically, they should at the household level. By setting up parallel structures without mapping out who’s doing what, we never have a sense of what happens at the household level. We have a public finance or budgeting system that is aligned with the idea of district development. But, we have not thought through the administrative architecture well.
“To get the district diversity right, we need clarity on the role of the districts and local governments. There is a need to fix our data systems by moving away from the information islands. And, finally, it’s time to make it political, where we get more and more people invested in the district as a unit,” says Avani Kapur.
“Any government has to fulfil economic and social objectives and the challenge is to align them. There has been development in aspirational districts and the disparity between districts is reducing. Critical gaps are being filled, yet, the challenge is to surpass districts outside the purview of the program of the aspirational district,” says Dr Amit Kapoor, Chairman, Institute for Competitiveness, Gurugram.
Dr Shreya Sinha, University of Cambridge, UK, says “Some regions have faced structural constraints and are not as agriculturally or industrially developed as others. There is a lot of conversation around the need for convergence in district development. This language of convergence and divergence is not typical to India and has been the parlance of development and modernisation theory in the post-world war years. This has been subject to a fair amount of critique.”
According to Sinha, divergence is kind of a catch-up discourse because it separates the developed and the underdeveloped as if they are completely unrelated and as though the development of one is not dependent on the underdevelopment of the other.
Probing from the point of view of social identities Sinha points, “a district which is dominated by Adivasis would have extremely different production conditions when compared to an aspirational district in Punjab. They can’t be seen through the same lens even at the national level.”
“There is a need to increase employment and encourage industry in aspirational districts. How can districts subject to jobless growth for the past two decades boost the economy? This is a political question. It’s about the lack of planning in any form. There are questions of federal structure, decentralisation, and resources. Hence, everything boils down to political will, whether it’s about data collection or devolution of power,” says Sinha.
Dr Jyotsna Jha, Director, Centre for Budget and Policy Studies (CBPS), Bengaluru, says that “It is not only about the size or level at which we want the data, but also about an administrative unit - a unit with some amount of independence to act. We have mostly uncoordinated data as we don’t want districts to become strong and independent. Even though we have governance structures, we do not have a matching system of data that would facilitate local-level decision making. That’s why it suits upper tiers - state governments and central government. And, hence the question of intent remains relevant.”
Although the quality of the government data for schematic reporting has improved as compared to 20 years ago, that does not necessarily mean that planning has improved. These are two different things. This data is just for reporting. Sometimes, even mandatory data is unavailable in the public domain. The data system will improve when we have a culture of transparency and a commitment to accountability, says Dr Jha.
Dr Jha highlights, “While structuring data for decision making, one needs to analyse how districts are feeding back data. Two districts with similar indicators can have very different kinds of needs. So, unless one links it to decision-making, the data systems are not going to change.”
The improved quality of data doesn’t mean the data speak with each other. These will end up as good data visualisations that lead nowhere. Data needs to be linked with a purpose such as decision making. It requires a good conceptual base and technological skills. The process and political stakes are as important given that there have been no action on numerous CAG, parliamentary and other committee reports flagging the implementation.
We have far more sophisticated ways of depicting data, but what we lack and perhaps where we have gone backwards is on linking data with decision making. It’s time to look at both data and diversity as political issues and not bring in technology as a sole enabler.
Manorama Bakshi is Senior Advisor, TATA Trusts and Arjun Kumar is Director, Impact and Policy Research Institute (IMPRI), New Delhi).