The Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS) entered India’s welfare architecture as something rare. It was not a dole or a discretionary programme, but a legal right to work. In theory, it altered the relationship between state and citizen. People demanded work; the state was obliged to provide it. Gram panchayats planned; administrations enabled. Wages were guaranteed, delays were penalised, and accountability was built in.
Nearly two decades later, the uncomfortable truth is this: For large parts of rural India, MGNREGS had long ceased to function as a right, even before the recent legal changes. What survived instead was a budget-bounded, administratively rationed, politically uneven scheme.
That distinction matters. Rights change power relations while schemes distribute benefits.
How a right quietly became a scheme
This dilution did not occur through any one dramatic policy reversal; it happened gradually. Registering demand for work, for instance, became a formality on paper, not a real reflection of people seeking work. This was evident in audit findings that showed demand for work was often not formally recorded: Applications were backdated, receipts were missing, and records were incomplete. As a result, ‘unmet demand’—the gap between households seeking work and work actually provided—became impossible to verify, even during periods of acute rural distress including in drought-affected states such as Rajasthan. This lack of verifiable data created space for a persistent claim that demand for MGNREGS work was low.
Employment came to be associated less with household claims and more with official announcements.
Similar shifts played out elsewhere. Planning templates replaced local priorities, wage delays weakened credibility, and gram panchayats increasingly implemented work they did not meaningfully design. Over time, employment came to be associated less with household claims and more with official announcements, marking a quiet shift from entitlement to expectation.
As a result, the right survived constitutionally but thinned institutionally. Only where district and local administrations were proactive, civil society organisations technically strong, or citizens politically organised did the original intent hold. One frequently cited example is Kerala, where state and local governments worked closely with Kudumbashree institutions to plan and execute MGNREGS works.
Studies have also documented pockets such as Mahendragarh district in Haryana, where proactive district leadership equipped panchayats to use the provisions for preparation and approval of the MGNREGS Labour Budget and annual action plan through the gram sabha, and parts of Badwani district in Madhya Pradesh, where sustained mobilisation by the Jagrit Adivasi Dalit Sangathan enabled workers to register demand and access employment.
But what explains the stark regional differences in how MGNREGS was implemented?
Divergent realities: Kerala, Odisha, and the limits of a rights-based programme
State-level evidence from MGNREGS reveals a politically uncomfortable pattern: The programme delivered its strongest outcomes not where deprivation was highest, but where administrative capacity and political will are stronger. The contrast between Kerala and Odisha makes this clear. The official Multidimensional Poverty Index (NFHS-5) shows that less than 1 percent of Kerala’s population is multidimensionally poor, while approximately 15 percent of Odisha’s population falls into that category. Yet Kerala consistently extracted far greater value from the employment guarantee.
Between FY 21–22 and FY 24–25, Kerala generated 62–68 days of work per participating household, and in recent years 20–26 percent* of participating households completed the full 100-day entitlement. This amounted to approximately 5.19 lakh families in FY 2024-2025. Women accounted for more than 80 percent of total person-days.
Odisha’s trajectory diverged sharply. Despite much higher multidimensional poverty, average employment typically remained at 50–57 days per household between FY 21–22 and FY 24–25*. Even in recent years, only 4–10 percent* of participating households completed 100 days of work. Women’s participation hovered close to 50 percent, shaped more by irregular work availability than enforceable demand.
A further distinction lies in payment processes. While delays persist nationally, Kerala generally completes key administrative steps more promptly, whereas in Odisha delays are more structural and multi-stage, diluting the credibility of timely payment as part of the right itself. This is not marginal. As of February 2025, unpaid wages under MGNREGS nationally exceeded INR 12,000 crore, underscoring how delayed payments have become a structural feature rather than an exception.
The implication is stark: MGNREGS did not automatically work better where need was greater; it functioned best where institutions converted a legal entitlement into lived reality.
This context matters when assessing the shift to Viksit Bharat-Guarantee for Rozgar and Ajeevika Mission (Gramin) or VB-G RAM G. Yes, the dilution of a rights-based framework is unsettling and, ideally, should not have been allowed. Rights once weakened are hard to reclaim. But for many rural households in Odisha, the change does not feel like the loss of a lived entitlement. That loss had already occurred years earlier. When work is intermittent, payments delayed, and planning distant, constitutional language offers limited daily protection.
Making the rozgar guarantee real in rural India: What must be done
If the original intent of MGNREGS had already ceased to operate as a right in much of rural Odisha and in other parts of India, the more relevant question now is whether the new framework can be shaped differently. VB-G RAM G will not succeed by virtue of being new. Its relevance will depend on how much it is embedded in local institutions and planning processes. Here are some suggestions:
1. Planning must be anchored by local institutions
Planning under MGNREGS was designed to be bottom-up, with gram panchayats at the centre. Yet, expenditure patterns show that a narrow and repetitive set of works has dominated implementation across states, including Kerala* and Odisha*, despite wide variation in local ecological and livelihood contexts.
For employment planning to reflect local priorities, it must be driven by village-level decision-making.
For employment planning to reflect local priorities, it must be driven by village-level decision-making rather than administrative convenience. Gram panchayat development plans (GPDPs) must avoid theme-based annual versions and instead be grounded in priorities articulated through village prosperity and resilience plans (VPRPs) and the gram sabha. Employment planning must be sequenced across seasons and anchored in water security, land productivity, forest regeneration, and local livelihood systems, ensuring that the resulting works are tailored to the specific ecological and livelihood needs of each village.
VB-G RAM G’s potential lies in strengthening VPRP-led planning and locally sequenced ecological and livelihood works.
2. Ecological and natural resource management must be prioritised
VB-G RAM G recognises ecological works as legitimate employment activities, but this will remain notional unless gram panchayats are enabled to identify, prioritise, and aggregate such works from village-level processes.
In both Kerala and Odisha, a majority of spending is concentrated in a small set of standardised, earthwork-heavy activities. In Kerala*, six such works account for 63 percent of expenditure, while in Odisha* six activities account for more than 64 percent, dominated by water absorption trenches and state-scheme housing. Nationally, the same pattern holds, with two-thirds of expenditure locked into a few categories.
This concentration favours works that are easily measurable, fit standard cost norms, are quicker to sanction, and are less likely to be questioned in audit than complex, site-specific ecological or livelihood interventions. As a result, ecological and livelihood priorities articulated through village-level processes are often not reflected in final GPDPs, as plans are consolidated around standardised, easy-to-execute works.
VB-G RAM G could address this by enabling science-based natural resource planning at village and gram panchayat levels, drawing on geohydrology, local meteorology, soil and land capability assessments, to identify works that respond to place-specific ecological constraints rather than generic templates. This will require building and legitimising local service providers with technical and knowledge capabilities to support villages and gram panchayats. In turn, gram panchayats must formally pay for such services as part of programme planning and implementation.
3. Work allocation and payment processes must pay attention to digital access and equity
Digital platforms are not neutral. In contexts of deep digital divides, they concentrate control among those with devices, connectivity, and procedural knowledge. Experience under MGNREGS has shown how this creates new forms of exclusion, limiting who can track payments, correct errors, or contest delays.
MGNREGS mattered not only because it promised work, but also because it briefly altered the relationship between citizens and state.
For instance, mandatory Aadhaar-linked systems—first the Aadhaar-based Payment System (ABPS) and now Aadhaar-based e-KYC with facial verification—have made daily work and wage eligibility contingent on successful technological compliance. As of November 2025, nearly 69 percent of all registered workers and 47 percent of active workers had not completed e-KYC, placing them at risk of being marked absent or denied wages. Administrative pressure to show that all workers have completed e-KYC or other digital processes further led to large-scale worker deletions: More than 27 lakh workers were removed in a single month during the 2025 e-KYC drive. This echoes the 247 percent spike in deletions during the ABPS transition, shrinking the reach of the programme while obscuring contestation and grievance redressal.
Technology, therefore, must support transparency and timely payments, and not bypass gram sabha processes or divert resources away from locally articulated needs. VB-G RAM G could ensure that gram sabha–approved work plans carry greater authority than dashboard-based compliance metrics by enabling offline access, time-bound corrections, and panchayat-level payment status updates. Without such safeguards, digital systems can become tools of informational exclusion and wage denial rather than instruments of accountability.
MGNREGS mattered not only because it promised work, but also because it briefly altered the relationship between citizens and state. That promise weakened long before the law changed. The challenge now is not to nostalgically defend a right that no longer functioned efficiently, but to insist that the new framework does not reproduce the same institutional failures. VB-G RAM G will ultimately be judged not by its legal form, but by whether it restores predictability, local planning authority, and administrative credibility to rural employment. Without that, it risks becoming another well-intentioned scheme—large in scale, limited in impact, and distant from the lives it seeks to secure.
*This data has been sourced and calculated based on analysis of publicly available data on MGNREGS in Kerala (a, b, and c) and Odisha (a, b, and c), accessed on 27 January, 2026.
—
Know more
- Learn about MGNREGS and how it was supposed to work.
- Watch this video explainer on the shift from MGNREGS to VB‑G RAM G. It outlines what the new law proposes and how it differs from the old framework.
