A Workable Poverty Line Is Still Elusive: Why Rural SECC Report Alone Won’t Be Enough

Reports that the NITI Aayog is planning to set up an expert group, to draw up a poverty line, is not only a commentary on the lack of clarity in the government on this issue, but also underlines how vexed it is. The Aayog had set up a task force on poverty soon after it was established, but that could not achieve any consensus on poverty estimation.

Separately, an expert committee headed by former finance secretary Sumit Bose on the Socio Economic Caste Census (SECC 2011) has reportedly suggested that data from this survey be used to deliver welfare benefits, instead of the traditional below poverty line (BPL).

Both these are important issues for the eventual rollout of a basic income transfer programme that the government is said to be mulling. Right now, all that is certain on this front is that the Economic Survey will have a chapter on the idea. If the basic income programme is targeted, and not universal, then estimation of poverty as well as identification of beneficiaries will be the key.

Does it really make sense, as the Bose committee has suggested, to junk the traditional poverty line for the SECC data, which captures a range of deprivations? And will it be more acceptable?

Actually, this debate emerged during the United Progressive Alliance (UPA) tenure when that almost needless controversy over the Tendulkar poverty line broke out in 2011. Completely ignoring the fact that the poverty line is established on the basis of monthly household consumption, this consumption expenditure was broken up into per person per day consumption, resulting in that much-ridiculed figure of Rs 32 and Rs 26 a day for urban and rural areas.

As part of the damage control, the government assured that the poverty line would be used only to estimate poverty at the macro level and that it would not influence the identification of beneficiaries for welfare programmes, whatever that meant. The SECC exercise then started and the rural SECC report came out in 2014.

This report gives an idea of not just income poverty but deprivation on several other parameters. It has an automatically included category – households that either don’t have shelter, or live on alms, or are engaged in manual scavenging, or have been released from bonded labour or belong to primitive tribal groups. It also lists seven criteria of deprivation – the condition of the house, no adult member in the working age group, female headed household with no male member in the working age group, households where the only adult member is disabled, no literate adult above 25 years, SC/ST households, landless households deriving a major part of their income from manual labour. There is an automatically excluded category – any household fulfilling any one of the 14 exclusion criteria.

To an extent, this is more robust and in tune with ground reality than the traditional poverty line, which is based on consumption expenditure of households. The threshold consumption expenditure is based on a certain assumption of what people need to meet their basic needs. Since it is normative, it is also subject to debate and controversy. The threshold is also based on household consumption surveys, which is based on a sample and is not a census of each and every household, as the SECC is.

The SECC data is also extremely granular, with locality and house number as well as details of family members, occupation, level of education, kind of house, ownership of selected gadgets, among other things.

But using SECC data to work out a poverty cut-off line will not be free of controversy. In the case of the rural SECC report, the automatically included category of 16.50 lakh households account for a mere 0.92 per cent of all rural households. Another 8.69 crore rural households (48.5 per cent of total rural households) fulfil one of the seven criteria of deprivation. A decision will have to be taken on whether all seven indicators of deprivation will be used to determine poverty (so only a family checking all the boxes will be entitled to welfare benefits) or just some, or a combination of some. So, any decision on this will entail a lot of political jousting.

Matters are complicated by the fact that only a provisional SECC report for urban areas is out; it has no mention of inclusion/exclusion criteria or deprivation indicators. If the SECC data is to be used as a measure of poverty, then urban poverty levels will end up not getting properly captured. Besides, some welfare schemes are common to rural and urban areas (a basic income certainly will be) and targeting these will be a problem. The government clearly needs to get its act together on the urban SECC report and finalise it as early as possible.

Is there a need, then, for a national poverty line? This is an academic exercise that has its own uses. Working out whether or not a new poverty line is needed and what it should be takes a long time. That can go on in a parallel track.