The Mahatma Gandhi National Rural Employment Guarantee Act: is it working?

  • Blog Post Date 29 November, 2012
  • Symposium
  • Print Page
Author Image

Jean Drèze

Ranchi University; Delhi School of Economics

Author Image

Neelakshi Mann

Ministry of Rural Development

Author Image

Varad Pande

Omidyar Network India

MNREGA is one of the government´s largest flagship schemes, and is the largest job creation programme of its kind in the world. Supporters believe that it is necessary to help rural workers smooth income in times of distress and increase labour market access for marginalised groups, whereas critics argue that it is taking labour from the troubled agricultural sector and doing more harm than good. What does the evidence really tell us - is MNREGA working or would resources be better spent elsewhere?

Comment on the validity of the following commonly held perceptions:

The Mahatma Gandhi National Rural Employment Guarantee Scheme (MNREGA) builds assets of questionable quality.

Jean Drèze

Almost anything is “questionable”, but if what is meant by this is that these assets are generally useless, I would not agree. The best MNREGA works have very high economic rates of return, according (for instance) to recent evaluations by the International Water Management Institute. The worst, I am sure, are useless – like ponds that are dug at the top of a hill, where water will never collect. I have seen many examples of both. Between these extremes, there is a whole spectrum of levels of usefulness, and the distribution of rates of return is anyone’s guess as things stand. We certainly need a lot more information on this. Meanwhile, the available evidence does not justify any sweeping condemnation of MNREGA works as “playing with mud”, as one commentator put it. In fact, several studies report that people’s perception of the usefulness of MNREGA works is generally quite positive. This is also what we found in a survey of about 100 MNREGA worksites in six north Indian states in 2008. The positive perception of MNREGA works was not just that of a vast majority of MNREGA workers, but also that of field investigators - 81%of the sample works were rated by them as “useful” or “very useful”.

Having said this, it is certainly true that the quality of a large proportion of MNREGA works is far from satisfactory, and that the quality standards could be very substantially improved with better technical support and other inputs. It is in this constructive sense that I would agree with the statement that MNREGA assets are of questionable quality.

Ashwini Kulkarni

Indeed, often the quality of assets does leave something to be desired, and it is not just because the assets are built using mostly unskilled labour; the casual attitude toward MNREGA projects of the engineers who design these projects is partly responsible.

However, I believe that the perception, especially of the urban elite, is a lot worse than reality. The community assets created by MNREGA do not make a visual impact like dams, flyovers, bridges and multi-lane tar roads. But they are nonetheless important for rural economic growth. The kaccha roads and watershed projects may not be visibly impressive but in order to understand their real value, just imagine the counterfactual. In this drought year in Maharashtra it is amply obvious that the villages that had done some watershed development projects did not face water shortage. This would not be the case if MNREGA projects were totally unproductive. And indeed it is cheaper to develop watersheds than to send water tankers to the villages!

Neelakshi Mann and Varad Pande

Field reports and independent studies suggest that at the micro-level, many productive MNREGA assets have been created on the ground owing to good planning and execution. However, significant challenges remain in bringing this to scale and there is a need for more focussed implementation with regard to the creation of durable and sustainable assets. Utility, sustainability, technical design and feasibility of the assets, are all important indicators to judge quality and impact. Studies on these suggest the huge productive potential of MNREGA.

Utility of the MNREGA assets have been confirmed through perception surveys of beneficiaries and stakeholders. One of the survey rounds of the MNREGA National Sample Survey Organisation (NSSO) in Andhra Pradesh, Madhya Pradesh and Rajasthan, found that around 99% of rural households in Rajasthan, 82% in Madhya Pradesh and 64% in Andhra Pradesh were using the assets created through MNREGA works 1. Further, out of all the MNREGA assets being used — 83% in Rajasthan, 80% in Madhya Pradesh and 67% in Andhra Pradesh — were considered to be of good or very good quality.

The sustainability of an asset depends to a large extent on the soundness of its technical design. A study evaluated 580 different types of works across four districts of Rajasthan and Madhya Pradesh, involving 640 households, with regard to critical design parameters under each category; for example, cost of investment, height, depth, technical design, quality of construction, time to recover the cost, among others. It was found that most of the structures, which have been built under the MNREGA, are sustainable and will last for their designed average life. In particular, wells, check dams and anicuts had been built with good quality material and the right kind of technical inputs. These structures could be sustained over a period of 10–15 years and through physical verification did appear sound enough to last that long.

A comprehensive study on Rate of Interest (RoI) (expected returns on the built asset) related to MNREGA works across eight districts of Bihar, Gujarat, Kerala and Rajasthan, is indicative of the productive potential of MNREGA. The study assessed 143 best performing MNREGA water related assets (for instance, irrigation, ponds, wells etc.) and found that RoI estimates are positive in the case of a majority of assets. Specifically, out of 143 assets2, 117 assets (for which detailed quantitative data on costs and benefits was calculated) had a RoI of over 100% in the first year, i.e. they recovered their investment in a single year of use. Across the four States, the RoI on all assets collectively was 126% for Gujarat, 121% for Bihar, 101% for Kerala, and 61% for Rajasthan. Micro-canal systems were found to have the highest rate of return compared to all other MNREGA works (more than 200% within a year).3 With renovation, water is available in these canals for up to eight months in a year and this has allowed the farmers to provide 3–6 additional waterings to their paddy crops. While the renovation increased the crop productivity by around 6-15%, the bulk of the benefits for the farmers came in the form of diesel saving as they were able to replace costly well-irrigation.


  1. National Sample Survey Organisation (NSSO), ‘Survey of MGNREGA 2010–11’, NSSO, 2011.
  2. It is important to note that the International Water Management Institute (IWMI)-Institute of Rural Management Anand (IRMA) study (Verma, S (2011), MG-NREGA Assets and Rural Water Security: Synthesis of Field Studies in Bihar, Gujarat, Kerala and Rajasthan, Anand: International Water Management Institute) chose only the best-performing water assets and therefore results are indicative of the potential of the Scheme. Thus, the study may not be representative of all MNREGA assets as the sample took into account only water-related assets and because it purposively chose only best-performing water assets.
  3. Verma, S (2011), ‘MGNREGA Assets and Rural Water Security: Synthesis of Field Studies in Bihar, Gujarat, Kerala and Rajasthan’, Anand: International Water Management Institute (IWMI).

Martin Ravallion

This is one of many poorly researched claims one hears about MNREGA. We have not yet seen the type of systematic survey data collection and analysis of the quality of the assets created that would be needed to properly test this claim. The bulk of the research on MNREGA to date has looked at targeting performance (“does it reach the poor?”), the direct impacts on poverty through the extra wage earnings of participants and the spillover effects on the wages for similar work in the casual labour market. We need more research on the assets created. My own anecdotal observations from my field work visiting work-sites in a few states suggest a mixed record, which leads me to caution against generalizing from such observations.

It can be agreed that MNREGA does not have asset creation as its sole, or even primary, goal. And the scheme faces trade-offs. The idea of the “employment guarantee” (since its inception in the early 1970s in Maharashtra) was in part to provide a means for vulnerable people to cope with risk. Almost everybody in rural India (including farmers with plenty of land) faces shocks that a scheme such as MNREGA can help them deal with. The beauty of the employment guarantee idea is that it acts like insurance, in that anyone who wants work at the going wage should be able to get it. This would have other benefits too, including empowerment. It would probably come at a cost, however. When an uncertain and sometimes large number of workers have to be accommodated on public works, the quality of the assets created might well decline. Meeting technical standards becomes more difficult. For certain types of assets, highly labour intensive production methods can come at a cost to their longer-term value. (The classic, though possibly over-used, anecdote is of roads that get washed away at each monsoon.)

In practice, however, there appears to be considerable un-met demand for work on the scheme. My research with colleagues at the World Bank indicates that roughly 40% of those people who want work on the scheme do not get it, and the rationing rate tends to be higher in poorer states—where it is needed the most. (See my paper with Puja Dutta, Rinku Murgai and Dominique van de Walle, “Does India’s Employment Guarantee Scheme Guarantee Employment?”) Naturally this undermines the insurance and empowerment benefits.

If MNREGA cannot do much better to achieve the promised “guarantee” then the cost-effectiveness calculations I have done point to the importance of creating assets that bring sufficient benefit to poor people. Only then will this type of scheme be likely to dominate over alternatives, such as a cash transfer scheme with the same budget outlay.

So my response to this claim depends to some extent on what measures are taken to assure that the current demand for work is met, especially in India’s poorer states. Without success on this front, asset creation for poor people should probably have higher priority if the scheme is to continue. However, it should also be acknowledged that many of the same factors in the local (village-level) political economy that have left un-met demand for work can also be expected to thwart efforts for asset creation favouring poor people. I return to these issues below.

In conclusion, while these studies point to the positive micro-level impact MNREGA assets have had, significant challenges remain in bringing this to scale. For instance, administrative issues like, shortage and quality of staff to prepare technical designs and take measurements, incomplete works owing to non-payment of wages, poor planning and/ or lack of demand still remain.

To address the issue of creation of durable assets, the next level of implementation of MNREGA, viz. MNREGA 2.0, has expanded the list of permissible works to strengthen the positive synergy between MNREGA, agriculture and allied rural livelihoods, to respond to the demands of the states for greater location-specific flexibility in permissible works and to help improve the ecological balance in rural India.

The Scheme’s new guidelines prescribe several measures to ensure timely and accurate measurement of work and payment of wages, including, deployment of additional human resources for timely measurement of works and record keeping. Cluster Facilitation Teams will be deployed in blocks where SCs & STs form more than 30% of the population.

MNREGA has been raising agricultural wages without there being any increase in labour productivity. This reduces farmers’ incentives to invest in their farms.

Jean Drèze

The validity of the claim that MNREGA has not been associated with any increase in labour productivity depends on the productive value of MNREGA works, discussed earlier. But let us suppose it is correct. Would that mean that MNREGA “reduces farmers’ incentives to invest in their farms”? Not necessarily, if only because many farmers have a lot to gain from MNREGA, both as part-time workers and as users of MNREGA assets (see below). Though little information is available on this, it is almost certainly the case that, overall, NREGA has boosted “investment in farms”, not undermined it. Just to give one example, in Jharkhand (one of the worst-performing states as far as MNREGA is concerned), nearly 80,000 wells have been constructed under MNREGA over the last few years, and many of them are very well built. The rural economy of Jharkhand, one of India’s poorest states, could literally be transformed through skilful MNREGA-enabled farm investments of this kind.

This does not apply everywhere, and there may well be areas where MNREGA discourages farm investments by raising agricultural wages without increasing labour productivity. Okay, so that would be one negative side effect to consider, along with many positive effects of MNREGA. This negative side effect needs to be seen in the context of the alarming stagnation of rural and agricultural wages in the period immediately preceding MNREGA. In the first half of the 2000s, when the Indian economy was enjoying runaway economic growth, the annual growth rate of real agricultural wages was below 2% according to National Sample Survey data, and close to zero according to more detailed information collected by the Labour Bureau. It is only after NREGA was launched in 2006 that the growth of real agricultural wages picked up again, though not dramatically by any means. Interestingly, this blip caused a mild panic in official circles and the business media (where expressions like “skyrocketing wages” were used with abandon) – a panic that contributed to the central government’s misguided attempt to delink MNREGA from the Minimum Wages Act in January 2009, but that’s another story. Anyway, my own question is – everything considered, is it a bad thing if MNREGA led to a moderate increase in agricultural wages, after a long period of stagnation? This increase was certainly a major gain for rural workers – not just MNREGA workers, but all rural workers, as a recent study by Clément Imbert and John Papp brings out. It would take strong evidence to persuade me that these gains are more than offset by the alleged losses of farmers who are being discouraged from investing or hiring. To my knowledge, no such evidence has been given so far.

Ashwini Kulkarni

There are two aspects to this statement: agricultural wages are rising, and farmers’ incentives to invest.

First, the perception about rising agricultural wages has been mostly voiced by large land owning farmers for whom it is important to not allow the bargaining position of the workers to move up. The impression conveyed through the farmers’ complaints is that MNREGA is a redistributive measure that merely transfers incomes from farmers to workers. This claim has been quite effectively challenged by Kanika Mahajan in her article “Agricultural Wages and MNREGA - Exploring the Myth” posted on 5th Nov 2012 on Ideas for India ( She traces the rise in wages to an improvement in yields during the period 2004-2009. The wage increase was correlated with productivity improvement rather than merely through a process of transfer from farmers to workers.

Second, given that the MNREGA projects need to fit into the 60:40 ratio of unskilled wages and other costs, the types of projects are limited - soil and water conservation, reforestation, roads and water storage structures. All these projects create the basic infrastructure for agriculture in that area. Small and marginal farmers who lack resources to invest in their own infrastructure stand to gain a great deal from the creation of such public goods.

Neelakshi Mann and Varad Pande

Most of the research studies have confirmed that an increase in MNREGA wages is one of the factors leading to an upward pressure in market wages.

MNREGA wages are linked to output and the increase in wages is linked to state-wise CPI (AL) (Consumer Price Index for agricultural labourers), it is meant to serve as a cushion for the Scheme’s beneficiaries and help them deal with inflation. Studies also indicate that MNREGA, by influencing wage rates in the rural unskilled labour market, has provided an additional opportunity for the Government to enforce statutory minimum wages.1 In fact, MNREGA has turned out to be one of the most effective ways of improving the welfare of the poorest. For example, based on NSSO 64th Round Survey during agricultural year 2008–2009, both male and female workers reported earning an average of Rs. 79 per day for work under the Act. These earnings are 12% higher than the average daily earnings for casual workers.2

It is difficult to indicate the precise impact of MNREGA on labour productivity in the absence of reliable data. In any case, as experts have argued, the solution to increase farm profitability lies in raising farm productivity to accommodate higher wages. The Planning Commission notes that there is preliminary evidence to suggest that with an increase in MNREGA, through better availability of water, land development, regeneration of natural resource base and efficient utilisation of MNREGA funds has led to visible growth in agricultural production and productivity. A study of 640 households in four districts of Rajasthan and Madhya Pradesh analysed the annual growth rate of agricultural productivity of non-irrigated land in 67 sample villages for 2006–2009.6 In three districts, the average annual growth rate of agricultural productivity was positive and ranged from 1% to 12 per cent across villages. This growth was directly attributed to the water-related works and structures taken up under MNREGA since the monsoon in the years of reference was below normal.


  1. Berg, E, S Bhattacharyya, R Durg and M Ramachandra (2012), ‘Can Rural Public Works Affect Agriculture Wages: Evidence from India’, WPS/2012–5, Oxford: Centre for the Study of African Economies Working Papers.
  2. C. Imbert and J. Papp, ‘Equilibrium Distributional Impacts of Government Employment Programs: Evidence from India’s Employment Guarantee’, Paris School of Economics, 2011.
  3. Institute for Development of Youth Women and Child (IDYWC), ‘Impact Assessment Impact Assessment of Mahatma Gandhi National Rural Employment Guarantee Scheme on Sustainable Asset Creation and Livelihood’, IDYWC, Report submitted to Ministry of Rural Development/UNDP, 2010.

Martin Ravallion

There is evidence that the scheme is putting upward pressure on agricultural wages, though this seems more pronounced in the states where the scheme is working better generally. (That is the main lesson I draw from the paper by Clément Imbert and John Papp, “Labor Market Effects of Social Programs: Evidence from India's Employment Guarantee.”) In the states where there is substantial rationing of the available work (and this is more common in the poorer states) I would not expect as much, or possibly any, impact on agricultural wages.

But it is hardly a bad thing if agricultural wages rates rise due to MNREGA. That will directly benefit many very poor people in rural India, given that there is a strong correlation between dependence on casual manual labour and poverty. (I calculate from the 20112012 National Sample Survey that the average participation rate in casual manual wage labour is about 50% for families in the poorest few percentiles of consumption, but it falls steadily as consumption rises, to close to zero for the richest.)

Nor should we presume that labour markets are working well in the absence of MNREGA. In fact, the existence of substantial involuntary unemployment/ underemployment is part of the rationale for the scheme. Thus a scheme such as MNREGA can help address this market failure that hurts poor people. Nor does the distortion have to be in the rural labour market, given that there is mobility to other (urban and peri-urban) labour markets; the unemployment could be in those markets, which generates pro-poor efficiency gains from MNREGA. (I showed this long ago, in a paper "Anti-Hunger Policies in Market Economies: Effects on Wages, Prices and Employment", in a volume edited by Jean Drèze and Amartya Sen, The Political Economy of Hunger, Oxford University Press, 1990.)

There is also a seasonal dimension to the labour market effects of MNREGA. The underemployment tends of course to be in the lean season. Workers benefit from the increase in employment over the whole year but also from its more stable spread. But many farmers will also benefit from this aspect of MNREGA in so far as it discourages out-migration in the lean season, which leads to reduced labour supply in the peak, given that some workers do not return. (This is a collective benefit that individual farmers may not fully appreciate.) MNREGA is more likely to even out the seasonal spread of employment and wages, which can be a benefit to the rural economy as a whole.

The effect on the investment by farmers is unclear. For one thing, that depends on how the scheme is financed. It would be good to see more research on that topic. What taxes rose, or what other spending declined, in response to MNREGA? We also need to spell out what counterfactual one has in mind. Is it the absence of MNREGA or some alternative use of the same public resources?

There is little doubt that MNREGA is building many private assets, such as wells that are largely or fully appropriated by individual farmers. I have seen examples in field work. (One might prefer the scheme to focus solely on public assets, or to secure financial contributions from the benefiting farmers. But that is not the point here.) This means that MNREGA is directly increasing farm investment — freeing up the resources of farmers to invest or consume.

Even if the scheme raises farm wages over the year as a whole it is questionable that this will reduce incentives for farmers to invest. Higher wages will induce a higher productivity of labour. (The marginal product of labour is determined by the wage rate.) The substitution away from labour to capital inputs will require higher investment. The asset creation from the scheme can also help stimulate private investment. Building collective irrigation channels, for example, will encourage individual farmers to make the extra investment to connect to those channels and adopt agricultural technologies and crops that require irrigation.

In many areas, there is no manifest demand for MNREGA projects from the potential beneficiaries. That may be due to the political pressure by the local elite (such as dominant farmers).

Jean Drèze

I think that the main reason for a lack of “manifest demand”, in the sense of formal work applications, is that formal applications are discouraged by the local administration. This is because a formal application creates a legal responsibility to provide work within 15 days, and potential sanctions (on concerned officials) in the event where this does not happen. The fact that very few states have even tried to put a formal application process in place is part of their general tendency to ignore or evade all the accountability provisions that were built into MNREGA. This underlines the need for an independent grievance redressal authority – going to court is not a realistic option for MNREGA workers, for obvious reasons.

Ashwini Kulkarni

The demand for MNREGA projects does not manifest for two reasons - First, the dominant farmers who are inevitably at the helm of village affairs suppress it for the fear that it would shore up the bargaining position of the workers (Anderson, Francois and Kotwal 2012), and secondly, the delays and the inefficiency observed in the implementation of MNREGA projects are discouraging to the potential demanders of work. Typically, work starts later than the stipulated period of 15 days. Wage payments come in much later than the stipulated 15 days and it takes much longer to start the next project. A series of delays of this kind discourages the labourers they either give up or opt to migrate. If the system improves and the delays are avoided, there would be greater demand.

Neelakshi Mann and Varad Pande

It is important to distinguish between no demand and rationing1 of demand for MNREGA work. “No demand” for MNREGA work may be a positive phenomenon, in fact it may indicate that the Scheme is realising its potential of making beneficiaries self-reliant. A rationing of demand, on the other hand suggests that there are households who are willing to work and seeking employment but not being given work.

Rationing of demand, wherever it exists, is no doubt a problem and greatly undermines the poverty alleviation potential of the Scheme. Rationing may be due to poor planning, that is, less availability of work, conscious stopping of works by state governments during certain seasons (agriculture season etc.), among other reasons. Influence of local leaders in allocation of work was also noted by a longitudinal study in five districts (480 villages) of Andhra Pradesh.The study found that, out of 4,800 households, 53% of the Poorest of the Poor (POP)3households and 56% poor households were JC holders. However, the actual participation rate (defined as the number of people who have worked in the Scheme) was only 17 per cent in Phase 1 districts, 11% in Phase 2 districts, and 5% in Phase 3 districts in 2008. The study found that the allocation of work was governed by leaders in the village. Further, the low participation of women and illiterates, indicated low awareness of the programme or some other constraints, like social factors etc.4

Rationing could also be due to a lack of awareness in the poorer states, or also due to a high demand and limited supply of work opportunities wherein women are forced to compete with men for employment, and the latter are usually favoured for manual labour.5

MNREGA 2.0 enlists provisions for States to efficiently capture demand thus reduce rationing. Some of key provisions that are being introduced in this regard are:

  • The GP or Programme Officer (PO), as the case may be, shall be bound to accept valid applications and to issue a dated receipt to the applicant. Refusal to accept applications and provide dated receipts will be treated as a contravention under Section 25 of MNREGA,
  • The provision for submitting applications for work must be kept available on a continuous basis through multiple channels so designated by GPs, these may include ward members, anganwadi workers, school teachers, self-help groups (SHGs) etc.
  • Provision must also be made (wherever feasible) for workers to register applications for work through mobile phones in addition to the MNREGA website and this should feed directly into the Management Information System (MIS),
  • The MNREGA software will automatically generate the pay-order for payment of unemployment allowance to such wage seekers whose demand for work is not met within 15 days (of demand). Reports prepared on this will be regularly tracked.
  • State Governments have to ensure that their monitoring systems capture even households that may have been purposely denied employment,
  • To estimate demand in advance, the district administration may conduct a door to door survey of Job Card (JC) holders.


  1. Dutta, P, R Murgai, M Ravallion and WV Dominique (2012), ‘Does India’s Employment Guarantee Scheme Guarantee Employment?’, Policy Research Paper, Washington DC: World Bank.
  2. Deininger, K and Y Liu (2010), ‘Poverty Impacts of India’s National Rural Employment Guarantee Scheme: Evidence from Andhra Pradesh’, Selected paper prepared for presentation at the Agricultural and Applied Economics Association, Colorado.
  3. The study defines Poorest of the Poor or ‘POP’ as those who can eat only when they get work and lack shelter, proper clothing, respect in society, and cannot send their children to school.
  4. Deininger, K and Y Liu, (2010), ‘Poverty Impacts of India’s National Rural Employment Guarantee Scheme: Evidence from Andhra Pradesh’, Selected paper prepared for presentation at the Agricultural and Applied Economics Association, Colorado, 2010.
  5. Dutta, P, R Murgai, M Ravallion and WV Dominique (2012), ‘Does India’s Employment Guarantee Scheme Guarantee Employment?’, Policy Research Paper, Washington DC: World Bank.

Martin Ravallion

To the contrary there seems to be a lot of demand! As I noted, our research suggests large un-met demand for work. I agree that the way local governance works is very important to the outcomes from MNREGA, but not so much through the demand for work. Rather it is the supply of work that can be reduced in local areas with poor governance. As long as it is feasible for local leaders to ration the work (leaving unmet demand), the number of project sites opened and (hence) level of jobs made available can be curtailed by high marginal costs facing the local elites in obtaining their share of the resources available. And increasing that marginal cost by tightening up regulations and so on can actually have a perverse effect of further increasing the un-met demand. The better way to attack problems of local corruption is to make it impossible for local elites to leave un-met demand for work on MNREGA. This will often require local civil society mobilisation, backed up by strong state- and district-level administrations to follow up on grievances.

There is a great deal of farmers’ opposition to MNREGA projects. In order to mitigate it, the government should use MNREGA money to subsidise wage payments on private farms.

Jean Drèze

I am not aware that there is “a great deal of farmers’ opposition to MNREGA”. The term “farmer” in this sort of statement actually refers to a minority of relatively large farmers who hire a significant amount of casual labour. Most farmers have a stake in MNREGA, partly as potential workers (since many “farmers” are employed at MNREGA worksites), and partly as users of the assets being created. These include not only community assets such as roads and ponds, but also assets associated with MNREGA works on private land, such land levelling, soil conservation, and the construction of wells. Some of the most productive MNREGA works are of this type.

Having said this, it is true that relatively large farmers sometimes oppose MNREGA, as has happened for instance in Andhra Pradesh and Tamil Nadu. But I don’t think that “subsidising wage payments on private farms” would be a good way of “mitigating” this, except possibly for specific and well-identified activities that farmers are unlikely to take up otherwise (land levelling is one example, which is already being implemented). The purpose of MNREGA is to create additional employment and earnings for casual labourers, not to foot the existing wage bill on behalf of the employers, with little employment creation. In any case, this idea is not practical.

Ashwini Kulkarni

MNREGA is a poverty alleviation scheme and has two objectives: (1) to create wage employment for the unskilled workers especially in slack season when none exists; and (2) to create productive assets such as roads, wells and watersheds that would help improve the agricultural productivity in the area. I believe that the policy measures should not deviate from these objectives.

I believe that it would be a mistake to appease farmers by giving them a wage subsidy. Yes, such a subsidy measure would result in fewer MNREGA projects being suppressed but it would shift the focus from helping the workers who have little or no work in slack season to helping the farmers whose demand for labour may rise only marginally. MNREGA is meant to provide ‘additional’ employment over and above what is available during the kharif. Thus, if the employment opportunity that is already available in kharif gets converted to MNREGA, there would be no additional income to the poor (workers). Also, if the government pays wages on private farms, it will remain a self-perpetuating system with no potential to wither away.

Farmers need to understand that MNREGA can create assets that can help improve productivity and that it has the potential to give a second season of farming to dry land regions. The Act does allow asset creation even on private farms. If the implementation of MNREGA projects, and consequently the quality of assets, improves the opposition of farmers would diminish.

Neelakshi Mann and Varad Pande

MNREGA was conceived as a social safety net for the poor and marginalised. The Act already provides for works such as irrigation, horticulture, land development on private land belonging to the SCs and the STs or below poverty line (BPL) families or to the beneficiaries of land reforms or to the beneficiaries under the Indira Awas Yojana (IAY) of the Government of India or that of the Small or Marginal Farmers (SMF). Around 12% of the total works in FY 2011–2012 alone were on lands of individual beneficiaries, including SMF. Permitting private works on lands of SMF implies a coverage of 40% of all cultivated area (80% of all land holdings).

Field reports indicate that due to an improvement in land productivity (particularly under the individual works taken up in MNREGA) and water security, small and marginal farmers previously working in MNREGA have gone back to farming. MNREGA may also be helping with crop diversification. The impact of MNREGA (as also mentioned above) is dependent on the planning and quality of assets created.

The opposition to MNREGA for large farmers is mainly on two accounts (1) MNREGA has increased wages so as to make farm labour unaffordable and (2) MNREGA is causing agriculture labour shortage.

As pointed out above, while MNREGA is increasing agriculture wages, this increase is directly linked to inflation and also ensures enforcing of a statutory minimum wage. Both of these have been overdue and ensure equity and may not have been possible in the absence of MNREGA. Alternately, the increase in cost of farming is also on account of increase in cost of agriculture inputs like water, fertilisers etc. which constitute the larger part of farm expenditure as opposed to labour wages.

It is argued that MNREGA is responsible for the labour shortage. However, there is no conclusive evidence to confirm this. In any case, at best, MNREGA is one of the many factors possibly influencing labour migration. Trends of a reducing agriculture force precede MNREGA; decline in agriculture labour, as a share of total economic activity, at the national level, is since 2004.1 Further, data from FY 2010–2011 suggests that 70% of the works in the Scheme have been generated during the agriculture lean season. A study also found that high non-farm wages have had a more significant role in the diversion of labour from agriculture than MNREGA.2

Overall, MNREGA has brought about greater inclusiveness and participation of the marginalised. It has provided a safety net for the most vulnerable. The Scheme also has the potential to encourage entrepreneurships among the marginalised through the increased household income and development of their land (taken up under MNREGA).


  1. Lanjouw, P and R Murgai (2008), ‘Poverty Decline, Agricultural Wages, and Non-Farm Employment in Rural India 1983–2004’, Policy Research Working Paper 4858.
  2. Indumatib, S and PM Srikantha (2011), “Economic Analysis of MGNREGA in the Drought–prone States of Karnataka, Rajasthan and Irrigation–dominated State”, Agricultural Economics Research Review, Vol. 24, pp. 531–36.

Martin Ravallion

As I have emphasised above, there can be benefits from such a scheme to many farmers (as well as workers), if the scheme works well. But no doubt some farmers will lose. For example, farmers in areas that normally receive seasonal migrants from other rural areas may well face labour shortages and higher wages, while farmers in the origin areas benefit. And it will be the losers that we hear complaints from. The government should then be cautious in its response, to properly weight gains and losses.

Are wage subsidies a better idea? The last time I looked into the literature it was not very supportive, though much of the research has been for developed countries. Argentina is one developing country that has also had a lot of experience with workfare schemes and has at times tried to develop wage subsidy schemes for private employers, as long as they employ workfare participants. 10 years ago, some colleagues and I did a randomized control trial of such a scheme in one urban area of Argentina. Some randomly chosen workers on the country’s workfare scheme were given the entitlement to a wage subsidy if a private firm employed them, with others retained as a randomised control group. (The experiment is described in my paper with Emanuela Galasso and Agustin Salvia, “Assisting the Transition from Workfare to Work: Argentina’s Proempleo Experiment,” Industrial and Labor Relations Review, Vol.57, No.5, October 2004, pp.128-142.). The results were not encouraging. Benefits were small, with very low take up of the subsidy by firms.

The Government of India should be open to considering options to all its programs, and MNREGA is no exception. But from what we know I am more inclined to think that some form of conditional or unconditional cash transfer scheme targeted to poor families has a stronger a priori case or consideration as an option to MNREGA, or possibly a complement to the program, notably when the transfers are also targeted to people who cannot or should not be taking up work on the scheme.

Information technology as well as the Unique Identification or Aadhaar (UID) can be used to reduce corruption in MNREGA projects.

Jean Drèze

Sure. They can also undermine the programme, if they are misused. Much depends on whether these technologies are deployed with workers’ interests in mind, or under pressure from commercial interests.

Right from the beginning, MNREGA has been the subject of many innovative applications of information technology. For instance, a unique “management information system” (MIS) for MNREGA, which involves pro-active display of all essential records on the internet (including wage payments down to the last worker and worksite), has been developed. This is a great monitoring tool, which is being continuously improved, and is now being extended to many other development schemes as well. In Andhra Pradesh, every step of the wage payment process is computerized, and in the event of delays (beyond the mandatory deadline of 15 days), compensation is automatically paid to the workers. This is a nice example of how information technology can be constructively used to further workers’ interests.

However, there are also examples of misguided attempts to force inappropriate or premature technologies on MNREGA. These include recent attempts to impose UID on MNREGA before the required infrastructure is in place. This would be a dream for the biometric industry, hardware vendors and related businesses. The UIDAI, of course, would also be delighted, since making UID compulsory for MNREGA workers would create a mass demand for UID enrolment. But this imposition could easily become a nightmare for MNREGA workers, judging from the recent experience of hasty transition to bank payments of MNREGA wages. This is not to deny that, if and when everyone has a UID number and the required infrastructure is in place (a big “if”!), there may be scope for some constructive applications of IUD in the implementation of MNREGA. But forcing premature applications of UID on NREGA is a recipe for trouble.

Ashwini Kulkarni

Certainly, Information Technology can play a big role. MNREGA website gives a glimpse of that. There are two ways in which this is possible. First, IT can be used for increasing transparency. Second, IT can speed up the whole process and reduce the delays in implementation and wage payments.

However, a number of improvements are warranted if ‘transparency’ and ‘speeding up the whole process’ are the goals. For greater transparency, the MNREGA website needs to be made more user friendly so that the NGOs or other stakeholders can download the data they require and use it to improve accountability. Second, an appropriate use of IT can bring about greater efficiency in all phases of implementation. The repetitive tasks can be made less cumbersome using IT; the measurements of the work performed by a team of workers (e.g. how much soil has been dug out) can be done by barefoot engineers with the help of better technology, thus obviating the need for engineers to visit the site every week. Once these measurement records are available regularly and timely then using the Measurement Book software can generate the Payment Sheet with speed and accuracy and this can then be shared with banks to make payments. Moreover, using IT automatically puts all data in public domain and time stamp of each task can provide the valuable information to the administration about the bottlenecks in the implementation cycle.

Delayed wage payment to the poor labourers is the most worrisome aspect of MNREGA implementation The delay occurs in every stage of the project: the evaluation of work done by the labourers; preparing payment sheet; depositing payments into the accounts of labourers through a post office or a bank. Of these, the first three reasons can be addressed effectively using IT, as explained in earlier paragraph. To ensure that the wages do reach the intended recipients, UID linked bank accounts can potentially do the trick. The Banking Correspondent model needs some kind of identification of the account holder and creating their biometrics requires additional investment for the Banks. Banks are reluctant to put in this kind of investment since these no-frill accounts of rural labourers’ funds’ float in and out of their account too fast for the banks to make any profit. Here the UID can provide the identification protocol and induce the banks to participate. Hence this route becomes beneficial and labourers can get bank accounts and banking services in the village with Banking Correspondent model.

Neelakshi Mann and Varad Pande

As part of the MNREGA next level implementation, the Government of India has already been placing a strong focus on the use of information communication technology (ICT) to promote timely wage payments and instill transparency at every level. At present there are 12 crore Job Cards and 9 crore Muster Rolls uploaded on the Management Information System (MIS); all MIS information is available in public domain. For FY 2011-2012, more than 90% of the total expenditure is available on MIS (including details of beneficiaries, wage payments, person days generates etc.)

ICT-based end-to-end solutions are also being piloted for capturing attendance, preparing electronic Muster Rolls (e-muster rolls), disbursing wage payments, etc. This will also enable real time data capture on the MIS and provide information at the panchayat, block, district and state levels. The government is also working with the States to move towards an Electronic Fund Management System (e-FMS) that will ensure timely availability and transparent usage of MNREGA funds at all levels.

Integration of MNREGA with UID through Aadhar is also being piloted in 51 districts across India1. Aadhar has the potential to institute efficiency and reduce leakages across the MNREGA life cycle including, capturing of real time attendance at work sites (which would lead to reduction in the number of ghost workers), reducing delay in wage payment, creating greater transparency and accountability in payments in general, etc. The Ministry has commissioned a study to analyse challenges, feasibility for scale up, where it is already being implemented.

Aadhaar in fact, has the potential to be superior to other biometric solutions for four reasons. First, it allows for interoperability among banks and Business Correspondents (BC), that is, the same Aadhaar biometrics can be used by any bank or BC that the worker may use. Second, it allows for uniformity of biometric standards across the country and across applications. Third, it is a single biometric service available across all government schemes and beyond, vitiating the need to do the biometric enrolment separately for different programmes. Fourth, Aadhaar is a mobile identity that travels with the resident even when he/ she moves or migrates.


  1. Ramesh, J, V Pande and N Mann (2012), ‘Aadhaar and MGNREGA are made for each other’, The Hindu, 4 July.

Martin Ravallion

Yes, I agree. But I would put greater emphasis on the importance of other reforms, as discussed below.

What would be your suggestions to make MNREGA more effective?

Jean Drèze

Here are a few quick suggestions (I have written about them in more detail elsewhere): (1) an independent grievance redressal structure should be created for MNREGA; (2) legally binding “rules” should be made to enforce the accountability provisions of MNREGA, including the unemployment allowance clause, the compensation clause and the penalty clause; (3) Schedule I of the Act should be amended to facilitate the creation of a skilled-labour cadre under MNREGA, so that the support staff (mate, Gram Rozgar Sevak, technical assistant, data entry operator, social auditor, among others) can be expanded under a well-defined legal framework, with cost-sharing between central and state governments; (4) MNREGA wages should be consistent with the Minimum Wages Act (suitably amended if need be), and indexed to the price level; (5) arrangements should be made for automatic compensation of MNREGA workers in the event of delays in wage payments, across the country; (6) one percent of MNREGA funds should be ear-marked for transparency measures; (7) the post of Programme Officer at the Block level should be a full-time post of rank equivalent to the Block Development Officer, in conformity with the Act and Operational Guidelines; (8) the Central Employment Guarantee Council should be a full-time body, independent of the Ministry of Rural Development; (9) an independent evaluation of MNREGA workers and worksites should be conducted every year under the auspices of this revamped Central Employment Guarantee Council; and (10) the MNREGA Guidelines should be simplified.

This list is based on the assumption that you wanted suggestions for public policy. For MNREGA workers, my suggestion would be that they form trade unions.

Ashwini Kulkarni

MNREGA needs to be made much more effective on both fronts -- namely, creating wage employment when there is no work, and productive assets generation.

For the wage employment objective to be fulfilled, the administrative system of implementation has to be sharpened. Andhra Pradesh has shown the way. I have been studying the AP systems for more than two years. First, they have an adequate number of personnel in place, especially at the levels of Mandal and Gram Panchayat; second, they have put a transaction based IT system in place from almost the beginning and they keep improvising and fine tuning it on regular basis; third, all their staff in the delivery mechanism (the cycle of starting the work, measurement of work, wage payments etc.) has had good training; the generation of regular, periodic monitoring reports and the authority responsible for monitoring and taking corrective measures is clearly laid out. Such a system ensures a smooth implementation that is lacking in various degrees in almost all other states.

The planning of ‘works’ to create productive assets is of immense importance. The Guidelines say that the Gram Sabha has to prepare a plan. However much we may love to have the Gram Sabha be so active, the collective action required for the same may not be possible in all villages. Two desirable developments that we would like to see are: (1) the Gram Sabha debating and planning ‘works’, and (2) Gram Sabha just picking from a given list of available ‘schemes’ as proposed by the State Government. The latter can at least ensure that all the schemes do not get dictated by the local official (e.g. sarpanch) implementing the program in that Gram Panchayat. This can serve as the first step and when some people in the Gram Sabha and or group of labourers get the confidence that their voice is heard then the next step would be to demand Works as per the need of the community and not rely on the menu card of schemes given by the State Government.

Watershed projects are important. But they must be based on meticulous and systematic planning of watershed treatments rather than the present practice of building ad hoc structures. A comprehensive watershed plan of each GP or micro watershed has to be prepared by involving various technical departments and the villagers. The list of ‘works’ emerging from this plan should be put in a chronological order and this can then become the ‘shelf’ for that Gram Panchayat. Other works like ‘roads’ can be added to this list.

The guiding principle has to be that the productive assets envisaged in the program provide the basic infrastructure for the livelihood of rural communities. By improving agricultural productivity while providing additional incomes to the poor, it should reduce the dependence of the rural poor on MNREGA. We have to make MNREGA work so as to make it wither away.

Further Reading

  • Anderson, S, P Francois and A Kotwal (2012), ‘One Kind of Democracy’, UBC Mimeo.

Neelakshi Mann and Varad Pande

The next level of implementation of MNREGA, viz. MNREGA 2.0 does take significant steps to detail and define roles, responsibilities and timelines at each level of the Scheme’s implementation to increase efficiency and effectiveness. The Scheme’s new guidelines attempt to institute reform measures to ensure effective & timely planning, timely wage payment, adequate shelf of works, streamlined fund flow, increased use of ICT, outlining procedures for social audits, for transparency. However, a larger part of the implementation of MNREGA continues to lie with State Governments. See

(a) Link to new works included in the MGNREGA Schedule,

(b) MGNREGA new Guidelines

For instance, the states could internally rank districts (based on indicators such as poverty ratio, demand, backwardness etc.) to identify blocks for focus of MNREGA. The kind of works to be taken up (based on watershed approach), deployment of human resources, monitoring systems (MIS, inspections, etc.) will need to be customised according to the area.

In six years of MNREGA, there are best practices that have emerged from states and districts (like Kudumbashree in Kerala and self-help-groups in Tamil Nadu for encouraging participation of women, social audits in Andhra Pradesh for improving transparency and accountability, project mode planning of work and convergence initiatives in Andhra Pradesh etc.- Ref. MGNREGA Sameeksha). These need to be shared and scaled up through districts or state initiatives.

Martin Ravallion

MNREGA is a complex scheme, and poses a number of challenges for effective implementation, especially in India’s poorest states, where the scheme is needed most but (as a rule) seems to be working less well, especially in meeting the demand for work. Yet the gains to India’s poor depend crucially on the scheme’s ability to respond when and where the work is needed.

In my view, making MNREGA work better will require a combination of strong civil society organizations and institutions (such as social audits) with more effective state and local public administration of the scheme. Let me elaborate.

Given the scheme’s current level of complexity, meeting the demand for work will probably not be possible unless poor states have greater administrative capacity in implementing MNREGA. Better performance will require better state and local-level administration, monitoring and reporting, including proper accounting for outlays and reporting to the centre. The shortage of the skilled manpower needed for these tasks is clearly constraining the ability of some of the poorer states to absorb funds and implement the scheme. Developing local administrative capabilities through training and incentivising staff, setting up strong financial management and monitoring systems is a high priority.

These enhanced administrative capabilities can also be used to assure local public awareness of households’ rights and of the scheme’s rules, local monitoring and grievance redressal, and to facilitate better local response to the expressed demand for work. A stronger civil society voice is needed, and this will require stronger state and local public administration. The aim here is to radically change the entire calculus of costs and benefits facing local leaders and officials. If workers and their communities know that they can demand work at the stipulated wage then they will resist any attempts by officials to ration them or take a cut from their wages.

It seems that there is still poor understanding of even the scheme’s most basic features amongst India’s rural poor. (For example, our surveys in Bihar indicate that too few people, especially women, currently understand that after getting a job card, they need to apply for work in order to get employment.) And civil society engagement in choosing projects and monitoring performance is often weak, especially in poorer states such as Bihar. This can change. My research on Bihar (with Dutta, Murgai and van de Walle) has demonstrated that a public information campaign using a movie can significantly enhance workers’ awareness. (We developed a fictional movie instrument to teach people their rights under the Act, and took it to a random sample of villages, with controls. Our forthcoming book, Rozgar Guarantee? Evaluating India’s Biggest Anti-Poverty Program in India’s Poorest State, describes the experiment and what we learnt from the results.)

This type of information campaign can also be used to help assure genuine participation in deciding what works are done, and assuring that they have lasting value to the community. However, such campaigns need to be combined with more effective implementation of the scheme’s supply-side provisions and a more rapid flow-of-funds. Then it should be possible to realise a greater share of the potential impact on poverty of this ambitious scheme.

It is important that reform efforts for MNREGA work on both these aspects—a stronger, more capable, local administration, plus more effective participation by civil society. One without the other will not assure a true Rozgar Guarantee.

No comments yet
Join the conversation
Captcha Captcha Reload

Comments will be held for moderation. Your contact information will not be made public.

Related content

Sign up to our newsletter