This structure should ring alarm bellsjust go back to credit ratings as a metaphor, especially municipal and mortgage-backed
securities. Intense reliance on simple summaries to direct money to complex entities and ends may be administratively appealing, but it is fundamentally risky. In this case, everybody loses from mistakes: the country loses some of the human and social capital needed for any version of development success; households miss out on the meagre support that social sector budgets are meant to be paying for.
We know that the prospects for a single perfectly accurate designation of the poor and only the poor are limited. India has been the centre of research on poverty measurement and identification for decades and theres some convergence, but nothing like a consensus. The methodology has changed dramatically every few years, each time under the watchful eyes of experts. The 1980 Integrated Rural Development programme used income, as recorded and reported by the village, to identify beneficiaries. The first BPL Census (1992) again used income, but replaced the guidelines for assessing it. The 1997 BPL Survey changed from income to consumption, from household to per capita metrics, and introduced a two-stage process to exclude some better-off families so that more attention could be given to ranking the poor. The 2002 survey switched the frame againto well-beingand included questions about assets, food security, livelihoods and labour market position to generate household scores that could then be converted into BPL/not-BPL depending on the number of poor allowed for each state. The present 2012 SECC again uses the two-stage model, but this time with inclusion as well as new exclusion criteria and new questions for the households in the middle.
And we know that the BPL designation can go terribly wrong and stay stuck there. The flaws in the scoring system used to identify BPL families in 2002 are enough to make one wonder what they were thinking. The scoring punished investment in latrines and childrens education, excluded people (at least in principle) if they had benefited from one rural housing scheme, reduced as many points for ownership of a radio or cooker as a TV, among other bloopers. It asked people how theyd like to receive assistance and awarded different points for the answers! The implementation was even worsethere were too few staff, the lists were revised after the fact as they passed through various levels and so on, as detailed in the Saxena Committee Reports trenchant summary. After the dust settled, a 2008 analysis by Himanshu published in Mint found that 61% of the people below the Planning Commissions (restrictive) poverty line did not have a BPL card, while 25% of households with consumption expenditure above the poverty line did. And yet, this designation stuck for a decade and may continue well into the next plan. Bihar, UP, West Bengal, Kerala, Tamil Nadu, and other states hadnt even started the 2012 SECC as of January.
We also know that India is a diverse country, where one size never really fits all. Some aspects of vulnerability are universalcertainly, living on alms should be considered poverty wherever one is (except in some temples, perhaps). But others, such as the significance of gender and caste and the economic interpretation of different forms of employment and size of landholdings, vary more. It is not impossible to incorporate diversity in a single BPL frame, but it also requires conscious effort and careful thought. We will soon see whether the Saxena Committees innovation of making landholding norms relative to local averages, relying on states designation of mahadalits, and encouraging states to add their own exclusion criteria to the compulsory list, will be enough to accommodate differences. Dr Saxena himself seemed concerned about how the weight given to landlessness and the SC/ST designation would play out in identifying comparable poverty in different settings in a January interview with the Financial Express.
Ive discussed some of the history of how this institutional constellation came about in a 2010 paper with TN Srinivasan (in India Policy Forum 2010-11). The more important question is how to start unraveling it, and in particular with this Budget, this Plan, and this years policies.
First, limit the use of BPL designation as the criteria for social support. There are compelling administrative and governance reasons today to have more programmes rely on the BPL designation rather than make up their own criteria. However, the costs of relying on this crutch rather than acting faster on some of the long-running conversations on administrative reforms and capacity-building are high and will increase as (we hope) social spending evolves to acknowledge and respond to a group with diverse economic, social, and service needs.
Second, consider all of the reasons for selective universalisation. Fiscal analysis through the narrow lens of annual programme budgets encourages an ideological contest between fiscal practicality and moral obligation. There is room for more engaged debate on the full social costs of hasty targeting. Limiting the Food Security Acts coverage, for example, may turn out to be penny wise, pound foolish if one considers the long-run public health savings and the demonstrable educational gains (aka, human capital returns) that come from well-fed children and mothers. This kind of comprehensive accounting could be a task for the proposed Independent Evaluation Office.
Third, create more institutional room for innovation in channeling social spending to beneficiaries: devolve either revenues or expenditure control while focusing intensely on transparency about how the funds are spent. Take advantage of the Twelfth Plan to finally consolidate some of the centrally sponsored schemes and move more funds into less restricted block grants. Federalism is supposed to be a policy laboratory and its not clear that citizen oversight will be worse than public sector monitoring of social programmes.
Movement on any of these fronts would start to open up new channels for social investment to reach its intended beneficiaries. Each of them may be individually imperfect; collectively theyll be safer than betting everything on the BPL.
Jessica Seddon is head of research for the Indian Institute for Human Settlements, Bangalore