Five heady years of MGNREGA (HT MINT, September 1. 2010)

Sunday will mark the fifth anniversary of the landmark National Rural Employment Guarantee Act (later rechristened the Mahatma Gandhi National Rural Employment Guarantee Act, or MGNREGA). At a time when the country is debating the contours of similar landmark legislation for food security (the National Food Security Act), it is time to evaluate the working of MGNREGA, initially implemented in 200 of the country’s poorest districts and expanded to all rural areas by 2008.

Going by the statistics available with the ministry of rural development, MGNREGA is the largest programme for providing employment in rural areas. Official statistics say that with 2.83 billion person days generated, 52.5 million households benefited from it in 2009-10, out of the 113.2 million households that had been issued job cards. Scheduled caste/scheduled tribe (SC/ST) households accounted for 51.2% of total person days, and women for 49%. These are impressive statistics by any standard. Nonetheless, these need to be verified against independent evidence.

Though numerous field and micro-studies are available, very few large-scale surveys existed on the functioning of MGNREGA until the National Sample Survey Organisation (NSSO) decided to include it as part of its official employment-unemployment surveys. NSSO has now introduced a separate activity code (42) for all those working as wage labourers under MGNREGA. The 64th round of the survey (2007-08), the latest available, is large enough to provide a reliable estimate of MGNREGA work at both the national and state levels, and is fully comparable with the quinquennial round of 2004-05—the year immediately preceding the implementation of MGNREGA.

First, the broad results: According to NSSO, total employment generated by public jobs programmes in 2007-08 was 1.02 billion person days, 71% of the official estimate of 1.44 billion person days. At an average wage rate of Rs78.91 per day, this implies that the total wage bill was Rs8,040 crore, 74% of the official estimate of Rs10,738 crore. These definitely show an improvement over the usual claim that only 15% of government money reaches the people. But even a 25% leakage in MGNREGA wage payment is cause for concern.

How does this compare with pre-MGNREGA days? Public jobs programmes in 2004-05 generated 240 million person days of employment. In comparison, employment generated increased by 4.2 times in 2007-08. However, the increase varies across states and districts. Compared with phase I districts (the poorest 200), employment increased by more than 5.4 times in the 130 phase II districts, and by just 1.9 times in phase III ones. Considering that the districts in phase I and II were among the poorest, this also suggests that the largest expansion in employment generation happened in the poorest areas. Since these districts are also the ones where the programme was initially implemented, this may also represent strengthening of the delivery system over time.

However, there is considerable performance variation across gender and states. NSSO shows only 38% female employment compared with the official claim of 43%. The 200 poorest districts showed a marginal improvement in female employment, from 28% in 2004-05 to 36% in 2007-08. It actually declined in phase III districts. NSSO’s estimate of SC/ST employment, however, was similar to official statistics.

In terms of state-wise distribution, four states accounted for 71% of total employment generated— Andhra Pradesh (28%), Madhya Pradesh (19.7%), Rajasthan (13.5%) and Chhattisgarh (9.5%). Andhra Pradesh itself showed a massive increase in employment from less than 10 million person days in 2004-05 to more than 270 million person days in 2007-08. The increase in Madhya Pradesh and Chhattisgarh was also impressive. On the other hand, Rajasthan, at 1.31 times, showed little improvement. Compared with these four, the poor states of Bihar, West Bengal, Assam, Jharkhand, Orissa and Uttar Pradesh together accounted for only 18% of the total employment generated.

These outcomes have had positive externalities on other areas. Not only have casual wages increased by 4.5% per year in real terms, MGNREGA has also led to a narrower gender gap in casual wages. As reported by NSSO, there is a negligible gender gap in public employment; but even in casual employment in other works, male wages were only 50% higher in 2007-08 compared with 58% in 2004-05.

It is obvious that in a short span of five years, MGNREGA has performed much better than what sceptics expected—in terms of the level of leakages, as well as inclusion for disadvantaged groups such as women and SC/ST households and poor districts. While these highlight the huge potential that the programme has, they also underscore its dependence on institutional structures and innovations in programme delivery, as in Andhra Pradesh. The huge gap across states in implementation is partly a result of governance. But it is also a reflection of political will.

Most sceptics agree that inflation and drought in recent years did not lead to massive protests partly because the rural population was insulated by MGNREGA’s safety net. Such claims need to be verified. But the last general election is evidence that the scheme has paid rich political dividends.

Himanshu is an assistant professor at Jawaharlal Nehru University and a visiting fellow at Centre de Sciences Humaines, New Delhi.