Jean Drèze, Reetika Khera & Siddhartha

Corruption can be eradicated from NREGA, and the way to do it is to enforce the transparency safeguards that are built into the Act and the guidelines.

The National Rural Employment Guarantee Act (NREGA), launched two years ago in 200 districts, is going through a critical learning phase. During this period there are bound to be many procedural problems, all the more so as the NREGA guidelines are very exacting. This does not detract from the fundamentally positive nature of this initiative, or from the possibility of making it a success. But it did give the Comptroller and Auditor General (CAG) good reasons to demand remedial action when the programme came under its redoubtable scanner.

Contrary to media reports, the draft CAG report on NREGA does not present much evidence of large-scale embezzlement of funds, nor does it conclude that NREGA is “a failure.” The report focusses mainly on procedural lapses, and constructive ways to address them. This is a useful wake-up call, just a few weeks ahead of the extension of NREGA to the whole country.

The question remains whether NREGA funds actually reach the poor. In this connection, we report here some findings of recent “muster roll verification exercises” coordinated by the G.B. Pant Social Science Institute, Allahabad. The survey teams consisted of carefully-trained students from Delhi University and elsewhere. The muster rolls were selected through random sampling and obtained just before the survey, leaving little scope for window-dressing. The investigators interviewed the labourers listed in a particular muster roll and asked them to confirm the details of days worked and wages earned.

The methodology of muster roll verification was developed in Rajasthan in the context of the right to information movement. This learning process was also an opportunity to develop a range of transparency safeguards for public works schemes (such as the pro-active disclosure of muster rolls, regular maintenance of “job cards”, and social audits). Many of them have been incorporated in the Operational Guidelines of NREGA, and even in the Act itself. There is a good deal of informal evidence from Rajasthan that these safeguards can go a long way in preventing corruption. We have reported some of our own observations on this elsewhere (The Hindu, July 13, 2007).

This new series of verification exercises started in May-June 2007 in Jharkhand and Chhattisgarh, in areas where we had found evidence of large-scale corruption just two years earlier, under the National Food For Work Programme (NFFWP). For instance, in the Surguja District of Chhattisgarh, there was virtually no check on the embezzlement of NFFWP funds at that time. The situation was so bad that one of us was constrained to describe NFFWP as a “Loot For Work Programme” (Times of India, July 2, 2005).

In the same district, we were interested to hear this year, from a wide range of sources, that the enactment of NREGA had led to a steep decline in the incidence of corruption. This was borne out by the muster roll verification exercises: in a random sample of 9 works implemented by gram panchayats, we found that 95 per cent of the wages that had been paid according to the muster rolls had actually reached the labourers concerned. A similar exercise conducted in Koriya, the neighbouring district, led to similar estimates of “leakages” in the labour component of NREGA — only 5 per cent or so.

In Jharkhand, detailed muster roll verification of NREGA works in five randomly-selected gram panchayats of Ranchi District suggested leakages of around 33 per cent. Clearly, this is totally unacceptable but even this high figure (one of the highest we found anywhere outside Orissa) would not justify the claim that the bulk of NREGA funds fails to reach the poor. Further, in Jharkhand too there was evidence of a gradual retreat of corruption compared with earlier years, when it was not uncommon to find that entire muster rolls had been manufactured from top to bottom.

Next we went to Tamil Nadu to participate in a social audit of NREGA in Villupuram District, conducted in July-August 2007. There we found much evidence of serious efforts to prevent the spread of corruption in NREGA. For instance, the government of Tamil Nadu has initiated an imaginative system of muster roll maintenance, whereby each labourer has to enter her signature or thumbprint in the muster roll every day by way of marking attendance. This ensures not only that the muster roll is available for public scrutiny at the worksite, as required by the NREGA guidelines, but also that large numbers of people actually see it every day. In this and other ways, much progress had evidently been made towards a leak-proof system. Unfortunately, it was not possible to quantify the leakages, as the Villupuram social audit did not include systematic muster roll verification exercises.

A brief follow-up visit to Andhra Pradesh enabled us to observe and appreciate various initiatives to prevent corruption in NREGA. For instance, the government of Andhra Pradesh has taken the bold step of paying all NREGA wages through post offices. This is an example of the “separation of payment agencies from implementing agencies,” recommended in the NREGA guidelines. This system virtually removes any incentive the implementing agencies have to fudge muster rolls, since the payments are beyond their reach. In addition, Andhra Pradesh has put in place a system of institutionalised social audits, involving routine verification of NREGA records through participatory processes. Judging from our brief visit, and from the social audit reports, these safeguards are quite effective. While various forms of petty corruption (such as bribes being taken by postmasters) have emerged from the social audits, there is no evidence of the sort of large-scale fraud that plagued public works schemes in Andhra Pradesh just a few years ago.

Rude shock

After these relatively upbeat discoveries, we had a rude shock in Orissa, where muster roll verification exercises were conducted in October 2007 for 30 randomly-selected worksites spread over three districts (Bolangir, Boudh and Kalahandi). The findings of this investigation have been reported elsewhere (The Hindu, November 20, 2007). Briefly, we found that Orissa had barely begun the transition from the “traditional system” of corruption in public works schemes (involving private contractors, mass fudging of muster rolls, and institutionalised kickbacks) towards a transparent and accountable system. The transparency safeguards had been sabotaged by vested interests and the system was virtually unverifiable. In Bolangir and Kalahandi, the infamous “PC system” (whereby various functionaries demand fixed percentages of scheme funds) continued and seemed to absorb around 22 per cent of NREGA funds. The silver lining is that even in this corruption-ridden region, there were many indications of positive change. As checks and balances are put in place, the system is becoming harder for vested interests to manipulate, and corruption is coming down. The clampdown on corruption has recently intensified, after Orissa earned a bad name for mass corruption in NREGA.

After this, it was refreshing (literally) to head for the hills of Himachal Pradesh in December 2007. A sharp contrast emerged there between the two survey districts — Kangra and Sirmaur. In Kangra, we found a remarkable culture of transparency in public records, including NREGA. The muster rolls and other NREGA records were usually available for public scrutiny at the gram panchayat office, often in computerised form. Further, with one major exception (Minta gram panchayat), there was an almost perfect match between the muster rolls and workers’ testimonies.

In Sirmaur, however, there was evidence of significant fudging of muster rolls. In some cases, this had been done (illegally) to augment the material component of the scheme, without appropriating any funds, but there were also cases where embezzlement was involved.

More extensive investigations are obviously required to consolidate these findings, but even small-scale surveys of this kind yield rich insights. All the relevant information, including muster rolls and verification sheets, is available on request. These records would be useful reading for those who believe that NREGA funds are systematically going down the drain. Equally enlightening are people’s testimonies. No sensitive person can fail to be moved by the words they have used to describe how NREGA employment helps them to live with dignity, feed their children and send them to school.

Behind the diversity of these findings is one overarching lesson. Corruption can be eradicated from NREGA, and the way to do it is to enforce the transparency safeguards that are built into the Act and the guidelines. Along with this, swift action needs to be taken whenever corruption is exposed. This is not the time for a loss of nerve.

(The authors are associated with the G.B. Pant Social Science Institute, Allahabad University.)