In an era of growing globalisation and rising inequality, the Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) stands out as a unique attempt to provide a social safety net via a massive public works programme. The government as an employer of the last resort is an idea that has existed in policy discourse around the world for much of the 20th century, gaining most currency during the Great Depression in the United States. However, MGNREGA takes this policy to a new realm because of its massive reach, universal nature, and its initiation during a period of rapid economic growth.
It is a good time to explore the reach and impact of MGNREGA. India Human Development Surveys of 2004-05 and 2011-12, organised by the National Council of Applied Economic Research (NCAER) and the University of Maryland, surveyed about 27,000 rural households in 2004-05, before the Act was passed, and in 2011-12 when the programme was implemented in all districts. Hence, it provides a unique opportunity to examine household well-being before and after the implementation of the Act.
Our scorecard on MGNREGA focusses on three issues: (i) The reach and targeting of the programme; (ii) Experience of the households that participated in MGNREGA; and (iii) Broader changes in the rural labour markets between 2005 and 2012.
The MGNREGA website claims that 500 lakh households — about 36 per cent of rural households — obtained employment from MGNREGA in 2011-12. Our household survey finds only about 25 per cent of the rural households participating in MGNREGA. Another independent household survey, the National Sample Survey of 2009-10, also finds about 25 per cent of households participating in MGNREGA.Well-targeted scheme
Regardless of the discrepancy between administrative statistics and actual usage, the programme is remarkably well-targeted. This targeting operates at three levels. At the village level, the uptake in villages with low levels of infrastructure is higher (28 per cent) than in villages with better infrastructure (21 per cent). It is more difficult to organise new programmes in more backward areas, so MGNREGA’s success in achieving this goal is quite remarkable.
At the household level, households from the marginalised communities — Dalits and Adivasis — are far more likely to participate in MGNREGA (36 per cent and 30 per cent respectively) than other households (20 per cent). At the individual level, older workers are disproportionately more likely to participate in MGNREGA than in the general labour force. Women, too, have higher participation rates; although only 29 per cent of all non-agricultural wage workers are women, 44 per cent of all MGNREGA workers are women.
However, even for those households doing MGNREGA work, the number of MGNREGA work-days is not very large. About 50 per cent of participating households work 40 or fewer days. MGNREGA administrative data shows that less than 10 per cent of the households complete their full 100 days; the India Human Development Survey (IHDS) data record about 15 per cent of the participating households completing 100 days. In the run-up to the election, the government has raised the limits for Schedule Tribes (ST) households living in forest areas to 150 days, from 100. But 85 per cent of the participating ST households and 95 per cent of all ST households have not exhausted their current limit of 100 days. When asked by IHDS interviewers why they had not completed the full 100 days, 75 per cent of the MGNREGA participants cited “No Work” as the primary reason. It would seem more important to focus on ensuring the full 100 days of work for everyone than to increase entitlement to 150 days.
For those households that participate in MGNREGA, the income from MGNREGA forms about 14 per cent of their total income. While the Act mandates payment in cash for people who are not offered work, we found few respondents knew about this provision and even fewer availed of it.
It is not clear whether MGNREGA is providing alternative sources of work or attracting people who were formerly underemployed or disguisedly employed. The IHDS data document an increase of just five days of work for men over a 12- month period in rural areas and four days for women. This is not a massive increase, suggesting that some of the MGNREGA work may have replaced rather than added to former work.
The IHDS also documents other changes in rural labour markets. Among workers, non-farm work has grown substantially while an exclusive agriculture focus has declined. The proportion of individuals who focus solely on agricultural activities— cultivation, agricultural labour, and animal care— has gone down from 51 per cent of men aged 15-59 to 35 per cent; for women the drop is from 84 per cent to 66 per cent. Much of this drop comes from changes in agricultural wage work and caring for animals; own-account cultivation is unchanged. While we do not know that MGNREGA caused these changes, the alternative non-farm employment is certainly part and parcel of broader changes in rural labour markets.Increase in daily wage
This declining agricultural employment has accompanied wage growth for daily wage workers, particularly agricultural labour. For male agricultural workers, daily wages in constant terms grew from Rs. 90 a day to Rs. 134; for male non-agricultural workers they grew from Rs.126 a day to Rs. 155. The growth for women agricultural workers was from Rs. 62 to Rs. 91 and for non-agricultural workers from Rs. 77 to Rs. 111.
These wage increases for women are particularly interesting. Historically, the lack of non-agricultural work has constrained women’s wages. If MGNREGA is in any way associated with the growth in women’s wages, this is a positive outcome. But these observations may also point to a real concern for farmers — a possible lack of availability of agricultural workers and high wages during harvest time. Rising agricultural wages for both men and women and simultaneously declining agricultural wage work suggest that it would be a sensible precaution to ensure that MGNREGA work is not timed for the peak agricultural periods.
The above discussion has noted several concerns with MGNREGA, particularly the discrepancy between official data and household reports on usage as well as the potential wage impact. But we have also noted that the programme has been particularly successful in providing employment to Dalits, Adivasis, and women, thereby serving as an attractive employer of the last resort to the most disadvantaged workers.
(Sonalde Desai, Omkar Joshi and Reeve Vanneman are with the National Council of Applied Economic Research and the University of Maryland. This is Part 2 of a five-part series on the NCAER’s new household survey. The views expressed are personal.)