Opinion Targeted welfare,universal attacks

Interestingly,economists on both sides of the delivery debate are united on the coverage aspect — both advocate almost-universal legal coverage. Their argument is that it is practically impossible to clearly identify different groups (BPL vs APL,or priority vs general) in a framework of differentiated entitlements. Experience suggests that a substantial section of the genuine poor (BPL or priority) gets excluded (exclusion error) while many non-eligible elements infiltrate (inclusion error) the group entitled for most subsidised items. Universal coverage would eliminate both errors,and is,therefore,the new consensus,despite the arguments over PDS or cash transfer/food coupons.

Apart from the exclusion/inclusion error issue,universal coverage has been a fixture of recent rights-based social welfare programmes. Right to work,under the rural job guarantee scheme,has universal provisions applying uniformly to all rural residents. Right to food should follow the same logic,rather than the differentiated foodgrain entitlements in the proposed law.

The NREGA experience is valuable in this context. While it appears to be a universal welfare regime for rural areas,it is actually a means-tested selective welfare programme for those ready to undertake unskilled manual work. A poor skilled artisan from a rural area is less likely to participate,since it is meant for unskilled manual work. However,in situations of desperation and livelihood insecurity,the same skilled artisan may participate. The NREGA,therefore,works through this self-selection of the poor,and ends up catering to the relatively worse-off within rural areas.

That prompts the question: should the NREGA not be used to identify the poor better? Despite the fact that,as many reports suggest,NREGA implementation has been plagued with corruption and mismanagement,it can be used to identify the poor provided this theoretical self-selection works real-time during its implementation.

The first panel survey on the NREGA,undertaken by the NSSO,which covered 82,108 people from 17,853 households drawn from 900 villages from three better performing states — Andhra Pradesh,Madhya Pradesh and Rajasthan — indicates that this self-selection is working. It was conducted between July 2009 and December 2009,during the 66th round of the NSS survey. Three findings of the survey support the self-selection theory. First,the monthly per capita expenditure of NREGA participants is “substantially less than” that of non-participants. Secondly,among those who have participated in the NREGA,the proportion of upper castes is less than 10 per cent,compared to those from scheduled castes,scheduled tribes and other backward classes. The social determinants of deprivation in Indian society have always indicated that upper castes are relatively better off,and the NSSO findings re-affirm the self-selection-under-NREGA theory along these lines as well. Third,the NSSO survey finds that household size for those participating in the NREGA is far greater than those of non-participants,in line with the idea that poor households tend to have larger families than those relatively better off. While the NSSO survey does not deal with the extent of irregularities in the scheme’s implementation,it certainly indicates that poor households are self-selecting themselves to participate in this means-tested welfare programme.

These findings leave little reason for planners to not consider the NREGA database to identify the poor. In fact,the Rashtriya Swasthya Bima Yojana (RSBY),the health insurance scheme for the poor,has already started using the NREGA database to identify a section of its beneficiaries. Any NREGA worker who has worked 15 or more days the previous year is eligible for health insurance under the programme. This may further be fine-tuned using state-specific cut-off days,given that poorly performing states like Bihar are likely to have fewer workers who have put in more than 15 days,because the programme has been relatively slow despite a large poor population. One can even consider district-wise cut-offs — say for the Kalahandi-Balangir-Koraput districts in Orissa or the left-wing extremism-affected districts.

Coming back to the debate over the extent of coverage under the food security law,the attack against the targeted approach is a disservice to needy sections,unless we are contemplating universal legal rights to decent clothing,housing,drinking water,electricity,higher education and health,etc,along with the universal legal right of foodgrains. Otherwise,targeted financial assistance (subsidy) from government needs to be supported and preserved as a mechanism. The targeted approach can also bring about a convergence of various welfare programmes for the identified group.

This means that identification of the poor will remain the most critical component. Rather than advocating that this targeted mechanism be entirely replaced by an “almost-universal” system,one should consider finding ways to improve identification of poor. For instance,as the government carries on the BPL survey,which is still to gather momentum,it could consider seeking information about NREGA participation in the household survey. This information would be a useful supplement for the seven other criteria being used by the government to identify the poor for various welfare programmes.

ravish.tiwari@expressindia.com