Limits of the Socio-Economic Caste Census 2011

Economic and Political Weekly 

It was the Ministry of Rural Development which, for close to five years beginning in 2010, designed, planned and oversaw the execution of the 2011 Socio Economic and Caste Census (SECC), whose first batch of results were released earlier this month. Yet, it was somewhat unusual to see Union Minister for Finance, Arun Jaitley, rather than his colleague in Rural Development, Rao Birender Singh, holding centre stage at the release event in New Delhi. This is enough of a clue about what the census is going to be used for.

The SECC has its origins, first, in the widespread dissatisfaction with the third below the poverty line (BPL) Census of 2002 and, then, with the former United Progressive Alliance government’s decision in 2011 to carry out a caste census, in a less than serious manner. With many central and state government programmes directed at the BP Lpopulation, surveys were periodically carried out by the states under direction from the Ministry of Rural Development to identify the potential beneficiaries of the schemes. The problem with the BPL surveys was that they had substantial errors of exclusion and inclusion (those who should have been identified as BPL were excluded and those who were not BPL were included, respectively). A more transparent criteria-based approach formed the basis for the 2011 SECC. Households which did not meet even one of the 14 criteria covering ownership of assets (a three/four wheeler, refrigerator, etc), regular employment (working in the government) and income (a household member paying income tax, etc) would be considered deprived. In addition, certain kinds of households—all manual scavengers, all ragpickers, etc—would automatically be considered as suffering extreme deprivation.

The new approach made more sense and the initial results available for rural India do confirm the continuing incidence of extreme deprivation; its most striking expression is that the income of the highest earning member in 75% of households is less than Rs 5,000 a month. Yet, none of the data tell us anything new unless we had been carried away by a mobile density of 90 per 100 persons as being an indicator of a very low overall level of deprivation. The information that has been thrown up by the 2011 SECC on the overall degree of deprivation is to be found in different forms in the National Sample Survey, Census of India and even in the, somewhat dated, National Family Health Survey.

If yet there is some excitement in the centre about the SECC it is because of what the government thinks it can do with the data. The 39% of the rural population identified as being deprived because it does not meet even one of the 14 criteria covered in the SECC is much lower than the 75% of the population that is to be covered under the National Food Security Act (NFSA). Could the central government then be thinking of cutting back on its food security obligations by restricting NFSA coverage to 40% of the population? The finance ministry may be attracted by that possibility but that would require the NFSA to be amended and the state governments to agree. The finance ministry may also feel that the availability of information categorising the entire population according to different kinds of deprivation would make a shift to cash transfers—as many in the government want to—much easier. Others see this as “big data” which can be used to track the population’s characteristics.

All this presupposes that the mass of information that has been collected is accurate and that the data has been validated. The initial analyses of the rural data throw up some anomalies. News reports point out, for instance, that according to the survey the incidence of ownership among households in New Delhi of fishing boats plying the Yamuna waters is higher than along Tamil Nadu’s long coastline. With all the data for all the districts yet to be compiled and the methodology of compilation yet to be examined, the validity of the 39% number as deprived cannot be assumed to be accurate. The finance ministry is therefore better advised to leave it to the rural development ministry to first put together the final results of the SECC for rural and also for urban India.

Caste is the other set of data which is yet to be released. We are now told that the Niti Aayog will compile and put out the information. Census information on caste should have been canvassed by the agency with the best skills for the purpose— the Office of the Registrar General and Census Commissioner, India (ORGI). However, the ORGI baulked at collecting this information as part of the Census of 2011. The government of the time then decided to conduct a separate caste census but in the end tagged it along to the rural development ministry’s socio-economic survey, which was conducted by individual state governments. It has been pointed out by a former census commissioner himself that a census on the complex issue of caste identity is not one to be tossed about from agency to agency. The caste data when it is finally put out may show up even bigger weaknesses of the 2011 SECC.

In the end, the SECC will turn out to be most useful only if the socio-economic component of the data is used for the specific purposes for which it was collected. The Indira Awaas Yojana could, for instance, cover only households living in one room or kutcha houses as identified in the SECC. It would be a case of overreach if the finance ministry sees the SECC as providing it an opportunity to reduce coverage and slash welfare expenditure.