In this article, Martin Ravallion, Professor of Economics at Georgetown University, contends that the main proximate reason for MNREGA's disappointing performance is that many people in poor areas of rural India who want work under the scheme have not been able to get it. To match the reality of MNREGA with its grand vision, poor people need to be made more aware of their rights and entitlements under the scheme, and the supply side needs to be more responsive.
Tweet using #MNREGA10yrs
The vision of the Mahatma Gandhi National Rural Employment Guarantee Act (MNREGA) has much intuitive appeal as an anti-poverty policy. Providing unskilled wage work on demand in rural areas will directly raise the incomes of poor families (since the non-poor are unlikely to want to do such work). There are other potential benefits, including:
- Insurance, to the extent that MNREGA work is available when it is needed.
- Empowerment, to the extent that people must take deliberate action to secure their rights under MNREGA.
- Public asset creation in poor rural areas, where there is often much to do.
- Second-round gains in unskilled wages, if the scheme is successful in soaking up the excess supply of labour at prevailing wage rates.
MNREGA's founders seem to have imagined, or hoped for, rather ideal local conditions for the implementation of their grand vision. The Act stipulates that everyone who wants work at the stipulated minimum wage rates for MNREGA should get it, up to 100 days per household per year; if not then an unemployment allowance is to be paid by the state government. Nor did the founders appear to expect that the scheme's direct participants would have to give up any other source of income to take up the work.
There has been some serious questioning of whether the ideal conditions assumed by MNREGA have held in reality over the last 10 years. My own thinking on this issue has been influenced by an evaluative study I did with colleagues at the World Bank. The study focused on Bihar where participation in MNREGA was surprisingly low given the high incidence of poverty. The calculations for our book, 'Right to Work?', showed that under ideal conditions the scheme would have brought down the poverty rate in rural Bihar by 14 percentage points or more. That would have been a great achievement. However, we also showed that the reality fell far short of this ideal. We estimated that the actual impact of the extra labour earnings from MNREGA in Bihar was a 1 percentage point drop in the poverty rate. This is clearly a disappointing performance for MNREGA in one of India's poorest states.
Extensive un-met demand for MNREGA jobs
The main proximate reason for the scheme's disappointing performance is that many people in rural India who want MNREGA work have not been able to get that work. There is extensive rationing of the available jobs across India, especially in poorer states. (And payment of the unemployment allowance appears to be a rarity.) That does not mean that the scheme is 'poorly targeted' — by which critics typically mean extensive leakage to the non-poor. Indeed, the evidence from 'Right to Work?' suggests that the scheme is reaching relatively poor families, even allowing for the un-met demand for work. Certain types of households are rationed, in that they want this work but do not get it. For Bihar we found that there is a clear gender dimension to the rationing, in that households with a large share of adult women and female-headed were less able to access MNREGA work when needed. In other respects, our research suggested that the rationing process in Bihar tends to favour those with characteristics typical of poor households. Those who are poor but lack the typical profile of the poor appear to be more likely to be excluded from access to the scheme when they want it. Among those who do participate, we also found a sizeable gap (10% or more on average) between the wages actually reported by workers and those they are supposed to receive under the scheme. So workers are not getting all the work they want, and they are not getting the full wages due.
Corruption and job rationing
Ironically, the efforts made to fight corruption on the scheme may well have generated some of the un-met demand we observe. While the total benefit to a corrupt local official rises with the amount of MNREGA work provided so does the expected cost, given the chance of getting caught. The local official's choice of how much work to provide balances the marginal benefit from corruption against the marginal cost. In a complex scheme such as MNREGA, those local officials who are tempted to take their cut probably expect that the marginal cost of corruption rises as the number of MNREGA projects and employment expand. This stems from the fact that most forms of corruption on this scheme require the participation of other agents, sometimes including the workers; as the scheme expands locally, the village leader must draw in and trust more people. The marginal benefit probably does not change much with the amount of NREGA work provided.
Thus, as long as rationing of work under MNREGA is possible, central or state government efforts to increase the marginal cost of corruption (such as by heavier fines or more surveillance) tend to reduce the amount of MNREGA work that is made available by local officials. To the extent that corruption is a bigger problem in poor states, we may then see why rationing is more common in those states.
Non-MNREGA work opportunities
The tightening we have seen in India's rural and urban unskilled labour markets over recent years has thankfully meant that there have been other opportunities for those families not finding MNREGA work. Some observers have credited MNREGA with the rise in private sector (including agricultural) wage rates over recent years (though with some signs of a slowdown of late). But that does not seem plausible, especially in states such as Bihar where so many people who have wanted MNREGA work could not get that work. The more plausible explanation for rising wage rates is the boom in the (non-MNREGA) construction sector, in both urban and rural areas. How long this will continue is unclear, and it remains the case that a great many rural families face serious downside risk, as evident in the rural distress in large parts of the country due to recent droughts. The need remains for some form of effective social protection.
Matching the reality to the vision
Realising the potential gains from MNREGA will require that the work is provided on demand. For this to happen, poor people will need to be much more aware of their rights and entitlements under MNREGA. This can be achieved with the right information campaign. Our research showed that an entertaining movie was effective in significantly raising awareness of rights and entitlements under MNREGA (as shown in Ravallion et al. 2015). Interestingly, for the poor in Bihar, the impact of the movie was largely confined to those who watched it — poverty in India (as elsewhere) often comes with social exclusion from the flow of reliable knowledge, which is something that advocates of better connectivity to information need to keep in mind (Alik-Lagrange and Ravallion 2016).
But better information alone is not sufficient. Deeper problems are found, especially in poor states. Reform efforts for MNREGA need to assure: (i) stronger, more capable, local administrations, and (ii) more effective and secure participation by civil society groups. Arguably both of these are also needed for successful rural development as a whole. In many respects, the challenges facing MNREGA are the challenges of development more broadly in India.
Given the scheme's administrative complexity it is probably not surprising that it is working less well in India's poorest states and regions — exactly where it is needed most. Poor states such as Bihar lack the administrative capacity for monitoring and reporting including proper accounting for outlays and reporting to the Centre. The shortage of skilled manpower constrains the ability of poor states to absorb local development funds, such as available through MNREGA. Greater effort is needed in training and incentivising staff, and setting up stronger (and smarter) financial management and monitoring systems; Dutta et al. (2014) make a number of specific recommendations. The flow of funds has been far from smooth or leakage-free, although there appears to have been some recent improvements in this respect. Here 'e-governance' initiatives can help reduce leakage, as shown for MNREGA in Bihar by Banerjee et al. (2015).
Enhanced administrative capabilities can also be used to assure local public awareness of households' rights and of the scheme's rules, and local monitoring and grievance redressal. The essential aim is to radically change the entire calculus of costs and benefits facing local leaders and officials. If workers 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. If an information campaign can be combined with more effective implementation of the scheme's supply-side provisions then it should be possible to realise a greater share of the potential impact on poverty of this ambitious scheme.
Options to MNREGA
Making the reality of MNREGA match the vision will not be easy, especially in India's poorer states. It is understandable that policymakers are considering options. One option is to reinvent the scheme as some sort of social fund for supporting rural development in lagging poor areas, with no pretense of an employment guarantee. Ideally the projects should be endorsed by local communities (at Gram Panchayats) and benefit the community as a whole, although these conditions are not easily met, especially in poor areas. The local authorities should be incentivised to find the cheapest way of delivering what is required; that need not be the most labour-intensive method. While attaining the employment guarantee would no longer be an aim, all the existing challenges of assuring that the asset creation happens, and benefits poor people, would remain.
Another option is a simpler transfer scheme, which does not impose a work requirement, and may only involve limited efforts at targeting poor people. For this purpose, one could imagine taking the MNREGA budget and giving a uniform cash transfer to all rural households in selected poor districts — a geographically-targeted basic income guarantee. Essentially, this would disconnect the work provision from the transfers under MNREGA, making it an easier scheme to operate in places with low administrative capacity. This too would face challenges, although the new personal identification system (Aadhaar1) could be valuable in assigning the uniform cash benefit. In the case of MNREGA in rural Bihar, and in past work on the original Maharashtra Employment Guarantee Scheme (Ravallion and Datt 1995), research suggests that if one had taken the scheme's gross budget and simply allocated it equally to everyone, whether poor or not, then one would have more impact on poverty than is achieved by the (self-targeted) allocation of net labour earnings from MNREGA. This second option would not attain all the potential benefits of MNREGA, but it would probably do a better job of reducing poverty in the short-term.
- Aadhaar is a 12-digit individual identification number issued by the Unique Identification Authority of India (UIDAI) on behalf of the Government of India. It captures the biometric identity – ten finger prints, iris and photograph – of every resident, and serves as a proof of identity and address anywhere in India.
- Alik-Lagrange, A and M Ravallion (2016), 'Social Frictions To Knowledge
- Diffusion: Evidence From An Information Intervention', NBER Working Paper 21877.
- Banerjee, A, E Duflo, C Imbert, S Mathew and R Pande (2015), 'Can E-Governance Reduce Capture of Public Programmes? Experimental Evidence from India's Employment Guarantee', 3ie Impact Evaluation Report 31, New Delhi.
- Dutta, P, R Murgai, M Ravallion and D van de Walle (2014), 'Right-to-Work? Assessing India's Employment Guarantee Scheme in Bihar', Equity and Development Series, World Bank.
- Ravallion, Ma and G Datt (1995), 'Is Targeting Through a Work Requirement Efficient? Some Evidence for Rural India', in van de Walle, D and Kimberly Nead (eds.), Public Spending and the Poor: Theory and Evidence, Johns Hopkins University Press, Baltimore.
- Ravallion, Martin, Dominique van de Walle, Rinku Murgai and Puja Dutta (2015), “Empowering Poor People through Public Information? Lessons from a Movie in Rural India”, Journal of Public Economics, 132:13-22.