SHANKAR ACHARYA: Guaranteeing jobs or fiscal crisis?

A PIECE OF MY MIND

Populism is abroad in the land. Free power for farmers is back. Subsidies on kerosene and LPG have increased. Special expenditure packages for particular states have been given. Price restraints on energy, steel and drugs seem back in vogue.

Public sector bank employees have got hefty wage increases. Schemes for debt write-offs are bandied about. The list grows by the week.

And this, despite many decades of development experience that clearly shows that populist policies cannot achieve rapid and sustained growth of incomes and employment for a country. On the contrary, such policies generally retard growth and poverty alleviation.

Perhaps the biggest joker in the current populist pack is the National Rural Employment Guarantee (NREG) Bill, which is to be brought before the next session of Parliament.

It could well be a time bomb that does irreparable damage to India's public finances and development aspirations. On the face of it, the Bill's goals are noble and reasonable [a draft is available on the website of the National Advisory Council (NAC) for the Common Minimum Programme (CMP)].

It says: "Every household in the rural areas of India shall have a right to at least 100 days of guaranteed employment every year for at least one adult member, for doing casual manual labour at the statutory minimum wage ..." Moreover, the accompanying explanatory note by Jean Dreze (not apparently a dreaded "foreign consultant") clarifies that the target population of the employment guarantee is poor rural households, who will "self-select" for the programme because of the "conditions of employment".

Dreze provides an estimate of annual costs. Each person-day would cost Rs 100, of which Rs 60 would be wages (approximately the weighted average of minimum wages for agricultural labour across states) and Rs 40 for materials and administration.

For 100 days this would mean Rs 10,000 per poor rural household. Since there are approximately 4 crore rural households below the poverty line (BPL), the total annual cost of the guarantee would come to Rs 40,000 crore.

That would be a hefty 1.3 per cent of current GDP, but allowing for a 4-year phasing in, it could start at 0.5 per cent of GDP and climb to 1 per cent by 2008/9. In the NAC/Dreze scheme, the wage component of the NERG would be borne by the central government and the material component equally shared by the Centre and states.

Thus, states would bear only 20 per cent of the costs and also the liability to provide unemployment allowance (at a minimum of one-third of the prevailing minimum wage) in case the administration fails to provide the necessary employment.

In each state, the programme would be administered by the district administrations together with gram panchayats, who will issue the necessary "job cards" to applicants who fulfil the eligibility conditions of rural residence, adult age, and willingness to do casual manual labour at the state's minimum wage for agricultural labour. (Note that being a member of a BPL household is not an eligibility condition.)

The crucial weakness of the scheme (and the costing estimates summarised above) is the presumption that only members of poor rural households will "self-select" themselves as applicants for the NREG. There are at least two reasons to seriously doubt this fundamental assumption. First, in many states people work at below the statutory minimum wage, which is rarely enforced under the ground realities of rural employment. So, many non-poor agricultural workers could switch from existing agricultural employment to the NREG.

Second, and more importantly, the scope for fake muster rolls and doctored job cards is enormous. With the job card as the key document for recording work done and claiming wage payment, it is easy to see that this piece of paper will become the key qualification for wage payment under the scheme, not any actual work done on a project site.

In most of rural India the reality of governance/administration may not distinguish between real and fake claims for work done under the NREG. In effect, the job card will quickly approximate a check for Rs 6,000 (Rs 60/day for 100 days) per year, with some discounts for appropriate "document processing".

If this is the reality, then every rural household will aspire to possess at least one job card. Dreze's presumption of "self-selection" by the rural poor could prove comprehensively false as could his estimates of the NREG's annual costs.

As Sunil Jain (Business Standard, October 25, 2004) has pointed out, the more relevant number for rural households eligible for the scheme then becomes 15 crore, not the 4 crore assumed by NAC/Dreze.

And the associated fiscal bill rises towards a mind-boggling Rs 150,000 crore per year, or about 5 per cent of GDP! (If one contemplates the prospect of two job cards per rural household, these numbers double!)

That would certainly destroy India's public finances and much else besides. Amongst other casualties would be growth in output and employment, including rural employment (of a real sort, not the likely fictions stimulated by the NREG), any semblance of moderate inflation and whatever limited capacity for development still possessed by the Indian state.

The costs of such possible breakdown would be most devastating for the rural poor, precisely the group that the NREG is aimed at helping! But then, the costs of well-intentioned but ill-conceived populist policies are often borne by the poorest in a society.

Confronted by such bleak prospects, what should we do? The most logical and first best option would be to do nothing, except perhaps focus on improving existing programmes of food for work and rural employment generation. Alas, politics rules over logic.

Given the commitments made by the UPA government, some version of the NREG appears inevitable. What can be done to contain the damage and improve the benefits from the programme?

The biggest incentive for leakage and abuse arises from the gap between the predominantly central funding of the scheme and the predominantly state/local responsibility for its administration.

One way of restricting an open-ended (and possibly ruinous) central liability is to cap it on a reasonable basis, such as by limiting central funding for the programme in each state to a product of its estimated number of rural poor households, its minimum wage for agricultural labour, 100 (days) and 0.8 (the Centre's share in total costs).

Any costs greater than this would be borne by the respective state, thereby providing a strong incentive to states to minimise leakage and abuse of the scheme. This would be consistent with the NAC/Dreze proposals.

Second, and for the same reason, the legal guarantee for the scheme should be given by the respective states, including bearing the responsibility for bearing the unemployment allowance (in case a state fails to provide employment to eligible applicants), as proposed by NAC/Dreze.

Third, to ensure states' commitment to the programme, it is important to maintain their share in the costs at the minimum of 20 per cent (corresponding to half the costs of materials and administration) proposed by NAC/Dreze.

Fourth, the scheme should be implemented in a phased manner over four or five years, allowing for modification from lessons learned.

None of this will guarantee the success of the programme. Nothing can. But without such reasonable checks there is a high probability of the ironic (and tragic) outcome that the finance minister who steered the nation out of a fiscal crisis in the early nineties might preside, as Prime Minister, over the genesis of the next fiscal crisis.

(The author is Member, 12th Finance Commission and a Professor at ICRIER. The views expressed are strictly personal)