On 1 July 2026 the Mahatma Gandhi National Rural Employment Guarantee Act, 2005 was repealed and replaced by the Viksit Bharat – Guarantee for Rozgar and Ajeevika Mission (Gramin) Act, 2025. MGNREGA had run for two decades as the closest thing India had to a universal rural safety net: any rural household could demand work, and the state was legally obliged to provide it within fifteen days or pay an unemployment allowance. The new Act keeps the guarantee and enlarges it. It also changes who pays for it, and that is the part with consequences.
The entitlement rises from 100 to 125 days per household per financial year, the wage floor is set at ₹300 a day, and the national average notified wage moved from ₹298.8 to ₹327.4. Those are real increases and should be recorded as such. The funding change is structural. MGNREGA was funded almost entirely by the Centre: it bore 100 per cent of unskilled wages and 75 per cent of material costs. VB-G RAM G is a centrally sponsored scheme shared 60:40 between Centre and state for most states, 90:10 for the north-eastern and Himalayan states and Jammu and Kashmir, and fully central only for Union Territories without a legislature.
That shift matters because of what the entitlement actually is. Under MGNREGA, spending was demand-driven: if more households asked for work, the Centre paid for it. Under the new Act, the Centre sets a normative allocation in advance and any expenditure beyond it falls on the state. An entitlement backed by a fixed budget is a different instrument from an entitlement backed by a legal obligation to fund whatever demand materialises — the guarantee survives in the statute, but its ceiling is now set by an allocation rather than by need.
States have said so directly. Punjab, Karnataka and Telangana are among those objecting to the 40 per cent share, and analysts have estimated state spending could rise four to five times. The Act also lets states notify up to 60 days a year as a pause covering peak sowing and harvesting, during which works are not undertaken. About five states notified July under that provision, which complicates any reading of the first month’s employment figures.
Those figures are contested, and it is worth being precise about why. The government has said 9.69 crore person-days were generated in roughly the first month, describing the transition as smooth. A July 28 release cited 5.2 crore person-days as of 22 July. Independent analysis has argued the reduction in employment against the previous July was very large, and has noted that the government’s dashboard does not carry a final July person-days report at all. When the headcount itself is disputed, the honest position is that the first quarter of this scheme cannot yet be assessed.
The larger point is that MGNREGA’s design was its politics: a legal entitlement funded on demand is expensive precisely because it cannot be capped in advance, and that was the trade-off the 2005 Act deliberately made. The 2025 Act has kept the words and changed the mechanism. Whether rural wages rise or fall under it will depend less on the 125-day figure in the statute than on how the normative allocations are actually set each year — and on what happens when a state’s demand exceeds its share.
