Learning from NREGA

Published: August 23, 2014 

Jean Drèze

Photo: The HinduFIGHTING FRAUD: Surveys point to a sharp reduction in the extent of embezzlement of NREGA funds in recent years, at least in the wage component of the programme. Picture shows NREGA workers Kancheepuram district, Tamil Nadu.

Corruption in NREGA works has steadily declined in recent years. There are important lessons here that need to be extended to other domains

One neglected aspect of the debate on the National Rural Employment Guarantee Act (NREGA) relates to the process aspects of the programme. In the process of planning works, organising employment, paying wages or fighting corruption, many valuable activities take place: Gram Sabhas are held, workers agitate for their rights, social audits are conducted, technical assistants are trained, administrators find out how to speed up wage payments, and so on. These activities, aside from being valuable in themselves, are also a great opportunity to learn.Prevention of corruption

One productive area of learning has been the prevention of corruption. The principal method of embezzlement in labour-intensive public works programmes is well known: muster rolls are inflated and middlemen pocket the difference. Before the Right to Information Act (RTI) came into force, muster rolls were beyond public scrutiny and the crooks had a field day. Things improved after muster rolls were placed in the public domain, and even displayed page by page on the internet. Even then, an enterprising middleman might fudge the muster rolls and hope that no one will bother to verify them. So, further safeguards were introduced one by one including mandatory social audits of all NREGA works.

A major breakthrough was the transition to bank (or post office) payments of NREGA wages. This was a painful affair — the system was not ready for it and the overload led to long delays in payments. Five years later, banks and (especially) post offices are still not equal to the task. For the prevention of corruption, however, this was a step forward: the new system makes it much harder to embezzle NREGA funds since the money now goes directly to workers’ accounts.

One major qualification is that village post offices are still vulnerable to capture by powerful middlemen. Extracting money from someone else’s bank account without his or her knowledge is very difficult because banks have strict norms of identity verification. But for a suitable commission, a village postmaster can often be persuaded to use the accounts of illiterate workers as a conduit to siphon off NREGA money. Over time, workers learn to collect their wages in person from the post office and verify the passbook entries. But it will take a while for many of them to protect their account from fraud. And the crook’s next refuge is to involve workers themselves in the scam.

How much progress has been made in this step-by-step battle against corruption? The second India Human Development Survey (IHDS) conducted in 2011-12 provides a tentative answer. Early tabulations of IHDS data, kindly shared by project director Sonalde Desai, suggest that 25 per cent of all rural households did some NREGA work in 2011-12. The average number of days of NREGA work was 49 per employed households, or 2.53 days per person for the whole sample. Multiplying this by the rural population total from the 2011 census yields an estimate of 210 crore person-days of NREGA employment in 2011-12. This compares with 219 crore person-days of employment being generated by NREGA in 2011-12 according to the Ministry of Rural Development. In other words, the bulk of official NREGA wage expenditure is fully reflected in this independent household survey.Survey findings

One survey is not conclusive evidence, but it certainly gives some reason for hope. The IHDS findings are consistent with those of another recent survey: the Public Evaluation of Entitlement Programmes (PEEP) survey. This was a relatively small survey (covering about 2,000 households), conducted in May-June 2013 in ten States: Bihar, Chhattisgarh, Himachal Pradesh, Jharkhand, Madhya Pradesh, Maharashtra, Odisha, Rajasthan, Tamil Nadu and Uttar Pradesh. In this survey, too, the number of days of NREGA work reported by the respondents (22 days on average, in 2012-13) was very close to the corresponding average for the same households (24 days) from official records.

The picture emerging from National Sample Survey (NSS) data is a little different. According to recent estimates by Clément Imbert of Oxford University, 68 to 78 per cent of official NREGA person-days of work are reflected in NSS data for 2011-12. The corresponding estimates for 2007-08 (prior to the introduction of bank payments of NREGA wages) are much lower: 42 to 56 per cent. Thus, NSS-based estimates of NREGA employment are consistently lower than the official figures, but the gap is narrowing over time. I suspect that the IHDS figures on NREGA employment are more accurate than NSS data because the collection of social statistics is one of the primary objectives of the IHDS survey, but not of NSS surveys. There are precedents of patchy collection of social statistics in NSS surveys, e.g. gross underestimation of the coverage of midday meals.

The good news is that each of these three surveys points to a sharp reduction in the extent of embezzlement of NREGA funds in recent years, at least in the wage component of the programme. However, it is not clear whether this also applies to the material component. Preventing an entrepreneur from submitting an inflated bill for materials is much harder than preventing fake muster roll entries. This problem, of course, is not specific to NREGA — it is endemic in the entire construction sector. This is one reason why the corporate sector loves government-sponsored “infrastructure” projects: there is plenty of scope for padding the costs.Productive value of NREGA

This point has a bearing on recent concerns about the productive value of NREGA works. There is a widespread belief that NREGA works can be made more productive by raising the material-labour ratio, because material-intensive works lead to the creation of tangible assets instead of earth structures that get washed away. This belief has no basis. It is all the more dubious, bearing in mind that material-intensive works are more vulnerable to corruption. Thousands of useless pucca structures have been built under programmes such as the Integrated Action Plan (IAP), Backward Regions Grant Fund (BRGF) and Member of Parliament Local Area Development Scheme (MPLADS) — more for the purpose of siphoning off material funds than to create productive assets. On the other hand, some labour-intensive works can be very productive, e.g. land levelling and contour bunding. Even a good earth road is often much better than a pucca road built with sub-standard material by a corrupt contractor. Judging from recent experience, there is nothing wrong with the current 60:40 norm for the labour-material ratio in NREGA works. Lowering the norm to 49:51 (an odd figure, perhaps borrowed from the stock market) would severely dilute the employment objective of NREGA without doing anything to make NREGA works more productive. A far better way of enhancing the productive value of NREGA works would be to provide more technical assistance to Gram Panchayats.

Much remains to be done to ensure that NREGA is corruption-proof — not just the wage component but also the material component. Meanwhile, the transparency safeguards that have been painstakingly built into NREGA are crying to be extended to other domains. In this and other respects, the programme is a great learning tool. This process aspect of NREGA deserves more recognition than it has received so far.

(Jean Drèze is visiting professor at the Department of Economics, Ranchi University.)

http://www.thehindu.com/opinion/op-ed/learning-from-nrega/article6342811.ece