When strategy or marketing teams think about how best to reach the Indian consumer, the big question is which unit should be looked at? Panchayat, city, constituency, district or state? The same question arises when we plan for investment allocations or develop program budgets. The answer often is a counter question: What do we have data for?
I argue that the NSS regional framework, as defined by the National Sample Survey Organisation (NSSO)—which divides India into 85 regions—offers a very useful lens for both business strategy and policy planning.
I bring this up in the context of the birth anniversary of P.C. Mahalanobis, 29 June, also celebrated as ‘Statistics Day’ in India.
The National Sample Survey (NSS), which was established in 1950 under his pioneering leadership, was intended to be a series of sprawling nationwide surveys designed to capture information on all aspects of the economic life of citizens. Over the years, it has provided detailed data on various economic, social and demographic indicators across India’s numerous regions.
The NSSO, for the purpose of its all-India surveys, divides each state into regions, each of which is a group of districts. As a result, the 28 states of India are subdivided into 85 regions.
How many people actually know about the NSS?: While everyone knows about the Census, the NSS seems to be largely mentioned only in circles of statisticians and researchers, and we hear of it when reports are released or when the veracity of public data-sets is questioned.
More than the data itself, let me draw attention to the NSS region framework that rarely gets discussed. Since 1950, expert statisticians at the NSSO—now a part of the Ministry of Statistics and Programme Implementation—have worked to arrive at a set of regions based on homogeneity related to agro-climatic conditions, demography, geographical contiguity, etc. This not only provides a meaningful sampling frame for the purpose of surveys, but can be considered for wider use to understand how Indian economic and social life is organized.. Which begs the question that when so much thought (and investment) has gone into developing this, why is it relegated purely to the realm of statistics and reports?
Let’s make the best use of what NSS has to offer: I propose that we extend the application of the NSS framework and actively use it for business growth, strategic planning and government policies.
Let me dispel the notion that NSS is for geeks. For example, businesses should consider organizing sales teams with the NSS lens and carrying out reviews at that level (a scientifically defined cluster of districts), especially for products and services that have significant consumer penetration at the sub-state level. Companies like Unilever and other FMCG majors have benefited from designing robust regional strategies based on slicing the market in ways that drive sales in a more focused manner. Not every company, however, may have the resources to invest in this inhouse.
Of course, the NSS approach is not the only form of classifying regions, but it does remain a well-developed frame that is available for wider use.
Similarly, strategy teams—at big businesses and new ventures alike—can benefit from analysing the economic potential of these regions and testing which ones are better suited as markets, raw material sources or avenues for growth and investment, and accordingly plan where to set up distribution channels, their next factory, etc, and formulate region-specific strategies.
Madhya Pradesh, for instance, is divided into seven NSS regions, Assam into four and Punjab into two, with each region bearing distinct features and its consumption expenditure patterns mapped out.
The NSS is versatile: From a government point of view, having states leverage the NSS lens to identify regional strengths will be relevant to come up with structural incentives and fiscal allocations tailor-made for specific regions within states, an approach that can add to both output and jobs. This also applies to philanthropies, civil society and anyone else working on grassroot development and service delivery. Designing and fine-tuning outreach efforts with the help of an NSS lens can offer a targeted approach that talks to the needs, aspirations and strengths of the locality and its people.
Steps can be taken to identify regions and clusters in need of focused attention to strengthen health and educational access, for example, or bring down multidimensional poverty.
If we look at the data itself or the information architecture, periodic NSS surveys provide region-wise estimates of poverty, employment, consumption and expenditure, to mention just a few indicators. There have of course been debates, but political will must join forces with measures to ensure the timely availability of unit-level data with relevant aggregates to enable not only academic, but also private-sector use.
Also, all other data-sets at sub-state level, whether generated by the government or by companies like Ola, Oyo, Swiggy, etc, should get aggregated at the NSS level and have NSS region markers. There are many known challenges in triangulating dispersed information. In fact, the designated NSS code should be made mandatory for all sub-state data-sets collected by any agency, public or private.
Layering this with the India stack and also newer data-sets like the GST network’s or satellite imagery can give us localized insights, but at a manageable level of aggregation.
India’s 28 states are too large and districts are too many (806 at last count). This NSS set of 85 regions may not be perfect, but are practical and representative for planning, governance, review and action.
Today’s India has much to thank Mahalanobis for, a statistician fondly remembered as ‘The Professor.’ Keeping his legacy alive means not just celebrating his work and the institutions he created, but also reimagining how essential building blocks like the NSS regional frame can be applied further.
I am cognizant of the challenges involved, but would argue that while we adopt newer tools and strengthen our digital public infrastructure, we can further accelerate this process by adapting, investing in and building on well-researched data frames that are (in a sense) public goods that exist but lie under-utilized. The wisdom of this great man—as it should be for us—lay in realizing that acquiring robust national statistics is not an end in itself, but a vehicle to enable proactive planning not just across governments at the central and state levels, but social enterprises and businesses as well.
The author is an impact advisor, former director Tata Trusts and author of ‘Regional Economic Diversity: Lessons from an Emergent India’