|Junk that SECC data|
|Saturday, 07 January 2017 00:00|
It overestimates poverty hugely, so is unaffordable
Though the central and state governments are believed to have agreed to start using data from the Socio-Economic Caste Census (SECC) of 2011 to guide their social-sector spending, according to a report in The Economic Times last week, they would be well advised to re-evaluate the plan. The advantage of the SECC, undoubtedly, is that there is a census-based list of people suffering from some kind of deprivation – once this list is populated with Aadhaar numbers, the entire social secotr spending can be better targeted. While updating the list is a big issue – a census takes 4-5 years to complete – a bigger advantage of SECC is that, since there are many deprivation parameters, each programme can choose their list of households; there are, for instance, 2.38 crore households that have one or less room and have kuccha walls and kuccha roofs.
A big problem with the data, however, is that it dramatically overestimates poverty. It is true, the old definition of poverty based on a recommended calorie count wasn’t the best, but the SECC data is quite at variance with what the NSS survey shows. So, for instance, the NSS showed 35% of households were engaged in casual labour while SECC puts the number at a much higher 51%. The SECC shows that 75.5% of households have an income of below Rs 5,000 (based on the income of the highest-earning member) while the NSS puts this at 62.8%. The NSS, as a result, has a mean household income that is around 45% higher than that in the SECC.
Not surprising, then, that the SECC has a significantly higher number of poor than India has ever had in its history, even if you use the higher World Bank parameters. In rural India, the SECC data puts the number of deprived households at 10.74 crore, of the total of 17.97 crore. It is true the number will be lower for households with different types of deprivation – 2.38 crore households with only kuccha houses will be the target for subsidized home-loan schemes, for instance – but if 60% of the population is to be considered deprived and social sector schemes have to be designed for them, the costs are going to be significant. In which case, the central and state governments need to think carefully if they want to use the SECC data. Also, with the central government now talking of a universal basic income – the chief economic advisor is on record saying this will feature in the Economic Survey – the costs of it can only be met by eliminating various existing schemes.