Success and failure in MGNREGA implementation in India

Download Success and failure in MGNREGA implementation in India

Preview text

ESID Briefing No. 1
Success and failure in MGNREGA implementation in India

• There is marked variation – both within and across states – in the provision of employment to those demanding MGNREGA work
• Employment outcomes are lower in states with higher poverty levels
• MGNREGA implementation depends on the supply of work rather than demand for it
• Supply is determined by differences in state capacity and commitment, local power relations, the role of lower-level functionaries, and political competition
• Improving employment outcomes requires strengthening the demand-driven aspects of MGNREGA through a focus on local level social audits, funding and tracking of outcomes
• States should be provided with support to improve their capacity and allowed a degree of flexibility in MGNREGA implementation

The Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) – offering up to 100 days work each year to rural Indians – is the largest social protection programme in the world, in terms of the number of households covered. There is a lively debate around whether MGNREGA provides a vital social safety net

for the poor or merely burdens the economy. Since 2006, it has expanded to cover all districts in India, providing work to 50 million rural households in 2012/13 at a cost of US$8.9 billion. But no one has satisfactorily explained why, despite similar implementation mechanisms, there is such unevenness in outcomes.

Success and failure in MGNREGA implementation in India

Figure 1 illustrates the dramatic variation in employment outcomes. In Andhra Pradesh, the average MGNREGA person days worked by


households that demanded work was almost twice that in Assam Differences in state capacity and commitment lead to variation

and West Bengal. There is also a large variation in the proportion in employment outcomes

of households demanding work that have completed 100 days of MGNREGA work. Further, one would expect demand for MGNREGA employment to be higher in poorer states. But Figure 2 shows that poorer states like Bihar and Odisha

There is no single route to better MGNREGA implementation; different states have emphasised different dimensions of capacity and commitment, with varying levels of success.

have provided less employment

Clearly, a state with high capacity – in terms

than richer states such as Kerala “There is no single route to better of its economic, organisational and human

and Tamil Nadu.

MGNREGA implementation”

resources, as well as its ability to reach out

To address this apparent paradox,

to civil society and potential beneficiaries –

researchers studied MGNREGA

is better positioned for success than a state

implementation in 10 states: Andhra Pradesh, Assam, Bihar, with low capacity. The same applies to its commitment, in terms

Chhattisgarh, Gujarat, Maharashtra, Odisha, Rajasthan, Tamil Nadu of initiative, preparation, mobilisation, sanctions, and continuity of

and West Bengal. They explored the relation between implementation effort. Capacity and commitment are interlinked and, by playing to

mechanisms and outcomes in terms of jobs generated.

its strengths, a state may be able to able to overcome weaknesses

in other areas.

Researchers compared 10 states at sub-national, block, Gram Panchayat and village level, using a combination of qualitative and quantitative methods including interviews, community meetings, field observations, document searches and the use of primary and secondary data for cluster analysis and cross-sectional econometrics.

For instance, Chhattisgarh is a poor state with low economic and organisational capacity. However, its strong commitment to MGNREGA at all levels has facilitated civil society involvement and led to high public awareness. This has boosted demand for work that in turn the state is committed to delivering. Poor states with similarly low capacity could look to improve their commitment at political as well as bureaucratic levels.

Figure 1. MGNREGA employment outcomes vary across states

Note: Averages across 2010-12. Source:; authors’ calculations.

Sustained engagement with civil society is critical. In Rajasthan, the early success of social mobilisation, demonstrated through extensive social audits, led to a backlash from Gram Panchayat heads (Sarpanchs) and ultimately the state. The subsequent clampdown on audits resulted in a loss of trust on the part of beneficiaries and civil society, and a sharp decline in both the supply and demand of MGNREGA work.

The high employment outcomes of

Andhra Pradesh, achieved despite “Sustained

minimal engagement with civil society, may suggest that a technocratic model is sufficient. But while bureaucratic

engagement with civil

autonomy, up to a certain point, is a good thing for MGNREGA implementation, Andhra Pradesh’s approach, which

society is critical”

deliberately bypasses Sarpanchs and relies

on state-mediated social audits without

sanctions, cannot be recommended. Poor states cannot afford

to suffer delays in federal funding as a result of not following the

rules; richer states should consider if, in achieving high employment

outcomes, they want to undermine other MGNREGA objectives such

as the strengthening of grassroots Panchayati Raj institutions and

social mobilisation.

Finally, prior experience in similar programmes has been widely assumed to aid states in MGNREGA implementation. But the poor employment outcomes in Maharashtra, where the employment programme that MGNREGA succeeded had very different objectives, suggests that this cannot be taken for granted. For a high capacity, committed state such as Maharashtra, the example of Tamil Nadu and Andhra Pradesh may be instructive. These states have used their capacity to provide more autonomy to bureaucrats, which has led to better employment outcomes.

Local power relations matter

Implementation is intensely political. At Gram Panchayat level in both Bihar and Gujarat, interactions among local groups and their political clout determined MGNREGA success. Where agricultural labourers had influence on the Sarpanch, employment outcomes were good. Where contractors had more influence, materials-intensive (as opposed to labour-intensive) projects were the norm. Labourers and marginal farmers preferred MGNREGA work when it did not clash with their farm work. But better-off farmers (with the exception of the wealthiest, who had other sources of income) tended to be opposed to MGNREGA. Farmers do not constitute a homogenous group in opposition to MGNREGA, and collaboration is essential to MGNREGA success.

MGNREGA work is driven by the supply of work made available, not the demand for it
Therefore state-specific capacity and commitment and stakeholder dynamics within states govern how much work is available, rather than the demand for it from rural households. Indeed, in Rajasthan only 4.57% of respondent households received work whenever they demanded it. Within states at the Gram Panchayat level, stakeholders form coalitions (or engage in ‘collusion’, whereby workers accept that a cut will be taken by those providing work), driving up employment outcomes. In this way ‘coalition’ as well as ‘collusion’ can be propoor. The supply-side nature of MGNREGA is well understood by beneficiaries: in Gujarat, MGNREGA is widely known as raahat kaam (literally, ‘relief work’), meaning it is viewed as a welfare measure rather than a guarantee of the right to work.
The role of lower-level functionaries could be critical to outcomes
The question of who governs the supply of work is a crucial one. A clear majority of villagers in Rajasthan perceived, correctly, that the type, allocation and implementation of projects within Gram Panchayats were influenced by the bias of Sarpanchs towards their own villages. So although almost 94% of respondents wanted more work, only half put in applications for it. At the block level, local political conditions affected funding for MGNREGA projects: in Rajasthan, the ruling party was channelling funds to its base in swing constituencies. Block and Gram Panchayat functionaries therefore have the potential to act as ‘valves’ to direct funds to certain constituencies; supply has to be ‘opened’ rather than demanded.
Political competition leads to better MGNREGA employment outcomes
In West Bengal, it was found that where there was political competition at village level, particularly between two parties, there was higher MGNREGA expenditure. This was also the case at block level in Rajasthan’s swing constituencies, with the ruling party’s fund disbursal to its base magnified to three times the norm. At state level, too, political competition boosted outcomes, though political fragmentation did not make for an effective polity. Political parties felt encouraged to facilitate MGNREGA in areas where they faced a challenge from other parties. Such an increase in expenditure, in turn, improved a party’s chances of getting re-elected. There is therefore a feedback effect on programmatic politics, indicating a shift away from clientelist, patronage-based politics.

Figure 2. Poorer states have worse employment outcomes

Note: Households (HHLDS) demanding work that have completed 100 MGNREGA days, averaged over 2010-12; Rural Head-Count Ratio (HCR) is for 2009-2010. Source: for data on households completing 100 MGNREGA days, and Himanshu and Kunal Sen (2013), Poverty in India: measurement, patterns, and determinants, mimeo.

These findings compel us to question some of the basic assumptions of the MGNREGA programme, in particular its demand-side nature. Some attribute the recent decline in employment outcomes to a decrease in demand stemming from the programme’s success in alleviating poverty. But they should consider whether the decline follows on from a decrease in supply of work – a gradual ‘closing’ of the valves. The findings also offer a more nuanced view of state capacity and commitment, as well as stakeholder dynamics, and their impact on employment outcomes. Those involved in implementing MGNREGA are recommended to:
Strengthen the demand-side nature of MGNREGA
The fact that supply has to be ‘opened’ implies that certain stakeholders have undue influence on the process of MGNREGA implementation, and may be actively blocking and diverting the supply of work. Currently, public tracking of MGNREGA projects stops at the Gram Panchayat level, making it difficult to assess if rural households are getting their fair share of work. In order to improve transparency and the accountability of Sarpanchs, it is recommended that MGNREGA projects be tracked right down to the village level. Demand at village level should be calculated to provide a better match with the nature and quantity of work provided.
There is also a strong argument for bypassing the discretion that district-level functionaries have over the disbursement of funds. The findings validate the Government of India’s decision to disburse funds directly to Gram Panchayats rather than to the district or block level through the electronic fund management system (EFMS).
Allow for flexibility – within limits
Federal support, over and above the guaranteed financial support to implement MGNREGA projects, is critical to improving state capacity and commitment. But in some cases a degree of flexibility may be needed to improve MGNREGA implementation. Preferably, states would petition the federal government for such allowances in advance. However, there has to be a strong commitment towards ‘non-negotiables’. Some of these conditions – providing employment, minimum wages and community assets, while ensuring that grievance addressal mechanisms are active – are mandated by MGNREGA itself. This study’s findings further emphasised the need to ensure that projects are demand-based, and that they are channelled through the Panchayati Raj institutions down to the village level.

ESID Working Paper 9: Delivering effective social assistance: Does politics matter? by Armando Barrientos and Sony Pellissery.
ESID Working Paper 15: The political economy of implementing the National Rural Employment Guarantee Scheme in India, by Abhiroop Mukhopadhyay.
ESID Working Paper 31: ‘They don’t want to work’ versus ‘They don’t want to provide work’: Seeking explanations for the decline of MGNREGA in Rajasthan”, by Deepta Chopra.
All available at
This brief was produced from an ESID project examining the political economy determinants of MGNREGA’s implementation. It was drafted by Rabi Thapa, with inputs from Dr Deepta Chopra (Institute of Development Studies, UK) and Professor Kunal Sen (University of Manchester). The research was undertaken by Dr Deepta Chopra, Mr Subhasish Dey (University of Manchester), Dr Himanshu (Centre de Sciences Humaines and Jawaharlal Nehru University), Dr Abhiroop Mukhopadhyay (Indian Statistical Institute – Delhi Centre), Dr Indrajit Roy (University of Oxford) and Professor Kunal Sen.

The Effective States and Inclusive Development Research Centre (ESID) is an international partnership of research and policy institutes led from the Institute for Development Policy and Management (IDPM) and Brooks World Poverty Institute (BWPI) at the University of Manchester. ESID is funded by the UK Department for International Development (DFID).
ESID researchers are based in Bangladesh, Cambodia, Ghana, India, Malawi, Peru, Rwanda, South Africa, Uganda, UK, USA, Zambia and elsewhere.
ESID is led by David Hulme, Chief Executive Officer; Samuel Hickey and Kunal Sen are Research Directors; Julia Brunt is the Programme Manager; and Pablo Yanguas is Research Associate.
DFID funds four Research Programme Consortia (RPCs) on governance and development, of which ESID in one. The others are the International Centre for Tax and Development (ICTD) at IDS, the Justice and Security Research Programme (JSRP) at LSE and the Secure Livelihoods Research Consortium (SLRC) at ODI.

Effective States and Inclusive Development Research Centre School of Environment and Development The University of Manchester Oxford Road Manchester M13 9PL UK
email: [email protected]
This document is an output from a project funded by the UK Aid from the UK Department for International Development (DFID) for the benefit of developing countries. However, the views expressed and information contained in it are not necessarily those of or endorsed by DFID, which can accept no responsibility for such views or information or for any reliance placed on them.

Preparing to load PDF file. please wait...

0 of 0
Success and failure in MGNREGA implementation in India