Follow us on:
Economy | June 2026
India feeds 81.35 crore people for free every month. It costs the central government Rs 2.28 lakh crore in FY26 (Revised Estimate), rising to Rs 2.27 lakh crore budgeted for FY27. That is more than India spends on health, education, and infrastructure combined in many years. The Food Corporation of India pays Rs 41.73 per kilogram to procure and distribute rice. It gives that rice away at zero cost. Every single rupee of that gap falls on the central budget. This article examines the full architecture of India’s food subsidy machine: who built it, who benefits, what it truly costs, where it leaks, and whether any government can afford to change it.
Rs 2.28L cr
Food subsidy Revised Estimate for FY26 was Rs 2,27,754 crore, a 12.1% rise over the Budget Estimate of Rs 2,03,000 crore. FY27 Budget Estimate is Rs 2,27,629 crore, a marginal decrease of 0.1% from FY26 RE.
81.35 Cr
Beneficiaries under NFSA/PMGKAY as of August 2025. Each receives 5 kg of wheat or rice free every month. Coverage is based on the 2011 Census and has not been updated.
Rs 41.73/kg
FCI’s economic cost of providing rice in FY26, including MSP, milling, storage, and distribution. Central issue price to beneficiaries: Rs 0. Gap per kg: Rs 41.73.
Rs 12 L cr
Estimated total cost of providing free foodgrains over the five years from January 2024 to December 2028, per government projections at the time of the PMGKAY extension announcement.
Table of Contents
- Part I: How India Built a Free Food Machine From the Bengal famine of 1943 to the PDS, NFSA 2013, and the PMGKAY extension to 2028
- Part II: The Money Trail Year-by-year subsidy data from FY20 to FY27, the FCI cost structure, and why the bill keeps rising
- Part III: The Food Corporation of India How India’s largest food procurement agency works, what it costs to run, and the Rs 29.80 per kg wheat paradox
- Part IV: Who Actually Pays? The taxpayer arithmetic, fiscal impact, off-budget borrowing history, and what the subsidy crowds out
- Part V: Leakages, Exclusion, and the Targeting Problem PDS diversion estimates, inclusion and exclusion errors, the 2011 Census problem, and Aadhaar-linked reforms
- Part VI: The Reform Debate DBT cash transfers, covering fewer people at a higher benefit, increasing the central issue price, and the political economy of change
- Frequently Asked Questions
Part IHow India Built a Free Food Machine
The Origins: Scarcity, Famine, and the State’s Response
India’s public food distribution system did not begin as a welfare programme. It began as a wartime rationing mechanism. The Bengal famine of 1943 killed between two and three million people. The colonial administration’s failure to distribute available grain equitably was a central cause of the catastrophe. The lesson that free markets could not be trusted to deliver food to the poor in times of scarcity was hardwired into independent India’s founding policy instincts.
The Food Corporation of India was established in 1965 under the Food Corporations Act, 1964. Its original mandate was price stabilisation: procure grain from farmers at a minimum support price to protect farm incomes, maintain buffer stocks to stabilise prices for consumers, and distribute through a network of fair price shops. This was not a zero-cost-to-consumer system from the start. Beneficiaries paid a Central Issue Price (CIP) for their grain, which was subsidised but not free.
The Targeted Public Distribution System, or TPDS, was introduced in June 1997. It replaced the universal PDS, which had been available to all ration card holders. The TPDS introduced a distinction between below-poverty-line (BPL) households, who got a higher subsidy, and above-poverty-line (APL) households, who got grain at a lower subsidy. The system was imperfect. State governments had wide discretion in identifying who was BPL and who was not. This created significant leakage and inclusion errors from the start.
-
1943Bengal Famine, 1943The Policy Foundation Is Laid
Two to three million people die in Bengal. The famine’s political impact shapes India’s founding food policy consensus: the state must directly manage grain procurement and distribution for the poor.
-
1965January 1, 1965Food Corporation of India Established
FCI is set up under the Food Corporations Act, 1964. Its mandate covers price support for farmers, buffer stock maintenance, and bulk allocation of food grains to state governments for distribution through the PDS.
-
1997June 1997Targeted PDS Introduced
The Targeted Public Distribution System replaces the universal PDS. BPL households receive grain at a heavier subsidy. APL households receive grain at a lower subsidy. State governments identify beneficiary categories, creating the conditions for widespread mis-targeting.
-
2013July 5, 2013National Food Security Act Enacted
NFSA transforms food entitlement from a welfare scheme into a legal right. Up to 75% of rural population and 50% of urban population are entitled to 5 kg of subsidised grain per person per month. Total coverage: 81.35 crore persons based on 2011 Census data. CIP fixed at Rs 3/2/1 per kg for rice, wheat, and coarse grains respectively.
-
2020April 2020PMGKAY Launched During COVID-19
The government launches the Pradhan Mantri Garib Kalyan Anna Yojana, providing an additional 5 kg of free grain per person per month over and above the NFSA allocation. Total grain distribution: 10 kg per person per month free of cost. Cost over 2020 to 2022: over Rs 3 lakh crore.
-
2023January 1, 2023NFSA Grain Made Free Under PMGKAY
The government merges the NFSA allocation and the PMGKAY free grain into a single scheme. The CIP for all NFSA beneficiaries is brought to zero. For the first time in the NFSA’s history, the statutory entitlement of 5 kg per person per month is provided entirely free of cost.
-
2024January 1, 2024PMGKAY Extended for Five More Years to 2028
The Union Cabinet extends the provision of free foodgrains to 81.35 crore beneficiaries for five years. Projected total cost over the five-year period: approximately Rs 12 lakh crore. This is the single largest social protection commitment in Indian fiscal history.
What the NFSA Actually Says
The National Food Security Act, 2013 is the legal spine of the entire food subsidy architecture. It covers up to 75 percent of the rural population and up to 50 percent of the urban population under the Targeted PDS. Total statutory coverage was fixed at 81.35 crore persons based on the 2011 Census and the Household Consumption Expenditure Survey of 2011-12. This figure has not changed since.
NFSA beneficiaries fall into two categories. Priority Household (PHH) members receive 5 kg of grain per person per month. Antyodaya Anna Yojana (AAY) households receive 35 kg per household per month. In January 2023, the Central Issue Price for all these categories was set to zero. Grain that previously cost Rs 3 per kilogram for rice, Rs 2 for wheat, and Rs 1 for coarse grains now costs the beneficiary nothing at all.
The legal permanence problem: The NFSA creates a statutory right. Once enacted, removing or reducing an entitlement requires amending the legislation. No government can simply issue an executive order to cut coverage or reintroduce a CIP above zero. The NFSA also specifies that any changes to issue prices must not exceed the Minimum Support Price. This legislative architecture makes cost containment politically and legally difficult. The 15th Finance Commission and multiple Economic Survey editions have called for a review. No government has acted on these recommendations.
Part IIThe Money Trail
Year-by-Year: How the Subsidy Bill Has Moved
| Year | Food Subsidy (Rs Cr) | % of GDP (approx.) | Key Driver |
|---|---|---|---|
| FY19 | 1,71,298 | 0.67% | NFSA baseline; CIP still non-zero; FCI off-budget borrowings distort true cost |
| FY20 | 1,08,688 | 0.38% | Decline partly due to shift of FCI borrowings off-budget; actual economic cost was higher |
| FY21 | 4,22,618 | 2.11% | COVID-19 PMGKAY launched; 10 kg free grain/person/month; also includes FCI off-budget borrowing settlement of Rs 3.43 lakh crore |
| FY22 | 2,87,194 | 1.02% | PMGKAY continued throughout; extended multiple times; remains at double the pre-pandemic level |
| FY23 | 2,87,195 | 0.94% | PMGKAY ended December 2022; but NFSA grain made free from January 2023; broadly flat |
| FY24 | 2,12,332 (RE) | 0.57% | NFSA grain free from January 2024; decline from peak due to end of additional PMGKAY 5 kg |
| FY25 | 2,13,000 (approx.) | 0.53% | PMGKAY extended five years; FCI economic cost rises; stable budget estimate |
| FY26 BE | 2,03,000 | 0.46% | Budget estimate at start of year; subsequently revised upward |
| FY26 RE | 2,27,754 | 0.52% | 12.1% rise over BE; higher wheat and rice procurement at increased MSP; OMSS costs; ethanol feedstock |
| FY27 BE | 2,27,629 | 0.6% (per Union Budget documents) | Broadly maintained at FY26 RE level; PMGKAY continues; 97% of DFPD budget allocation |
Why did the bill jump in FY21? FY21 is an outlier not just because of COVID-19 but because the government also settled Rs 3.43 lakh crore in off-budget borrowings by the Food Corporation of India (FCI). In prior years, the government had been under-provisioning the food subsidy in the budget and asking FCI to borrow from the National Small Savings Fund (NSSF) to fund the gap. These borrowings accumulated off the government’s books. In FY21, the government cleaned up this accounting fiction by formally taking these liabilities onto the budget. The one-time settlement made FY21’s food subsidy number look enormous. The underlying recurring cost without this settlement was in the Rs 2.8 to 2.9 lakh crore range.
Food Subsidy Trend: FY22 to FY27 Budget Estimate (Rs Lakh Crore)
Source: Union Budget documents, Department of Food and Public Distribution Demand for Grants, PRS India. BE = Budget Estimate; RE = Revised Estimate.
Why Does the Cost Keep Rising?
The food subsidy is the product of two moving parts. The first is the economic cost of procuring, storing, and distributing each kilogram of grain. This rises every year because the Minimum Support Price for wheat and rice is revised annually. In FY26, the economic cost for rice reached Rs 41.73 per kilogram, up from Rs 39.80 in FY24-25. For wheat, it was Rs 29.80 per kilogram in FY26, up from Rs 27.70 in FY24-25.
The second moving part is the Central Issue Price, which is what beneficiaries pay. Since January 2023, this is zero. So the subsidy per kilogram equals the full economic cost. For rice, that is Rs 41.73 per kilogram. For wheat, it is Rs 29.80 per kilogram. There is no mechanism to reduce this gap except by raising the CIP (politically impossible in the short term) or reducing the economic cost (which requires systemic procurement reform).
FY26 Cost Per Kilogram: Economic Cost vs Central Issue Price
Rice (economic cost)
Rs 41.73
Rice (CIP charged)
Rs 0
Wheat (economic cost)
Rs 29.80
Wheat (CIP charged)
Rs 0
Source: FCI Annual Report data; PRS India Demand for Grants Analysis 2025-26. Economic cost includes MSP, milling or handling charges, storage charges, and distribution overhead.
Part IIIThe Food Corporation of India
India’s Grain Giant: Scale and Structure
The Food Corporation of India was incorporated on January 1, 1965. It is a statutory body under the Food Corporations Act, 1964 and operates under the Ministry of Consumer Affairs, Food and Public Distribution. FCI handles over 70 percent of all food subsidies. It is responsible for procurement, storage, movement, and distribution of food grains. It operates through a network of over 2,500 godowns spread across the country, with a total storage capacity of over 482 lakh metric tonnes as of November 2025.
FCI’s role in the procurement cycle is central. It announces minimum support prices for wheat and paddy in advance of each sowing season. Farmers bring their produce to procurement centres. FCI and state agencies buy at MSP, regardless of market price. The grain is then cleaned, milled (in the case of paddy into rice), stored, and transported to state distribution points, from where it moves to Fair Price Shops and reaches beneficiaries.
Storage Capacity
482 lakh MT
Total capacity as of November 2025 across 2,500+ godowns
Grain Stock (Jan 2026)
58.40 MT
Central pool rice and wheat stocks vs buffer requirement of 21.41 MT
FCI’s FY26 Revised Est.
Rs 1.7 L cr
FCI’s own revised expenditure estimate for FY26, up from Rs 1.4 lakh crore
Rice Economic Cost
Rs 41.73/kg
FY26 estimated economic cost per kg including MSP, milling, storage, and distribution
Wheat Economic Cost
Rs 29.80/kg
FY26 estimated economic cost per kg including MSP, storage, and transport
Fair Price Shops
5.43 lakh
Total FPS across India; 99.6% are now ePOS-automated for biometric authentication
The Surplus Stock Problem
India’s grain stocks have been persistently above buffer requirements. As of January 1, 2026, central pool stocks stood at 58.40 million tonnes against a buffer and strategic reserve requirement of just 21.41 million tonnes. This is nearly three times the mandatory minimum. The surplus creates two problems.
First, storing excess grain costs money. FCI pays for rent, fumigation, staff, and maintenance at its godowns. Grain that sits in storage for months degrades in quality. The longer it sits, the higher the carrying cost. Second, the surplus puts pressure on FCI’s cash flows. In FY26, FCI had to take short-term loans of Rs 25,880 crore despite receiving over Rs 75,921 crore in central subsidy releases, because the timing mismatch between procurement outflows and subsidy reimbursements left it periodically cash-short.
The government has attempted to liquidate surplus stocks through the Open Market Sales Scheme (OMSS), ethanol blending, and the Bharat Rice initiative. In FY26, FCI sold a record 5.63 million tonnes of rice through various channels including OMSS and ethanol feedstock supply. This helped contain the FCI’s financial exposure but added to the subsidy bill because grain sold at or below market price for ethanol represented an implicit subsidy to the blending programme.
The on-budget vs off-budget distinction that matters: For several years before FY21, India’s officially reported food subsidy was significantly lower than the true economic cost of the food security programme. The reason was that the government was instructing FCI to borrow from the National Small Savings Fund to finance its working capital deficit, rather than drawing on the central budget. By FY20, FCI’s outstanding NSSF borrowings had reached Rs 3.43 lakh crore. In FY21, the government settled this entire amount through the central budget, causing the FY21 food subsidy figure to spike to Rs 4.22 lakh crore. Since then, the government has committed to full budgetary provisioning to FCI, making the reported subsidy number more accurate. The FCI is also now not expected to borrow to fund its subsidy shortfall.
Part IVWho Actually Pays?
The Taxpayer Arithmetic
The food subsidy is funded entirely from the central government’s budget. There is no cess, no dedicated fund, and no user charge recovery. The entire Rs 2.27 lakh crore in FY27 comes from general tax revenues, primarily from income tax, corporate tax, GST, customs, and excise duties.
India’s total tax revenue in FY27 is budgeted at Rs 44.04 lakh crore. The food subsidy alone consumes approximately 5.2 percent of gross tax revenue. The three major subsidies together, which include food (Rs 2.27 lakh crore), fertiliser (Rs 1.71 lakh crore in FY27), and petroleum (negligible at present), total approximately Rs 4.54 lakh crore per budget documents, or about 10.3 percent of gross tax revenue.
| Spending Head | FY27 Budget Estimate (Rs Cr) | % of Total Expenditure |
|---|---|---|
| Food Subsidy (PMGKAY/NFSA) | 2,27,429 | 4.3% |
| Fertiliser Subsidy | 1,67,000 (approx.) | 3.7% |
| Interest Payments (largest single head) | 13,90,000 (approx.) | 24.4% |
| Defence Capital + Revenue | 7,85,000 (approx.) | 16.0% |
| Health (Union Budget allocation) | 99,858 | 2.2% |
| Education (Union Budget allocation) | 1,39,000 (approx.) | 3.3% |
| Rural Development | 1,86,000 (approx.) | 4.1% |
| Capital expenditure (infrastructure) | 12,22,000 (approx.) | 24.5% |
The comparison is instructive. The food subsidy alone is larger than the entire Union Budget allocation for health. It is larger than the allocation for education. It comes close to matching the rural development budget. Yet unlike capital expenditure, which builds assets, the food subsidy creates no lasting physical infrastructure. It redistributes current production to current consumption. This is not an argument against food security spending. India’s extreme poor genuinely depend on this grain. It is an argument for understanding what the spending does and does not deliver.
The Fiscal Impact on India’s Deficit
India’s fiscal deficit target for FY27 is 4.3 percent of GDP. Total budgeted expenditure is Rs 53,47,315 crore against total receipts (excluding borrowings) of approximately Rs 36.52 lakh crore. The food subsidy contributes to a structural rigidity in the expenditure side: it cannot be cut without legislation, it rises automatically with MSP increases, and it is insulated from the normal appropriations process that allows discretionary spending to be trimmed in lean years.
What does Rs 2.27 lakh crore look like in per capita terms? India’s 2011 Census recorded approximately 121 crore people. Using this base, the food subsidy works out to approximately Rs 1,881 per person per year across the entire population. For a four-person household, that is Rs 7,524 per year. For comparison, the direct benefit received by each PMGKAY beneficiary is 5 kg of grain per month, or 60 kg per year. At the economic cost of Rs 41.73 per kg for rice, that is Rs 2,504 per beneficiary per year. The government is spending approximately Rs 2,504 per beneficiary in economic cost terms to deliver a benefit worth Rs 2,504 at economic cost, but worth only Rs 480 at prevailing market wholesale prices for rice of approximately Rs 30 to 35 per kg in most Indian mandis. The subsidy is large in fiscal terms but moderate in its private consumption value to each individual beneficiary.
Part VLeakages, Exclusion, and the Targeting Problem
The Scale of PDS Diversion
Studies and committee reports have consistently estimated PDS leakage at 25 to 40 percent of allocated grain. Leakage refers to grain that is allocated to the PDS system but does not reach genuine beneficiaries. It is diverted to the open market by corrupt fair price shop owners, stolen during transit, or misappropriated through ghost ration cards. The National Sample Survey Organisation and various parliamentary committee reports have documented this problem over decades.
The Government of India’s Aadhaar-linked PDS reforms have reduced leakage significantly. By August 2025, over 99.8 percent of ration cards had been Aadhaar-seeded. Over 99.6 percent of Fair Price Shops were operating with ePOS machines requiring biometric authentication before grain is issued. The One Nation One Ration Card programme allows beneficiaries to access their entitlement from any FPS across the country, reducing the lock-in that previously allowed dealers to ration or withhold grain.
Ration cards issued to households that do not exist, have migrated, or have died are called ghost cards. Before the Aadhaar-linked de-duplication drive, ghost cards were a primary source of PDS leakage. FPS dealers would draw grain against ghost card allocations and sell it in the open market.
The technology-driven reform process identified and cancelled crores of ineligible or duplicate ration cards. By early 2025, all states had completed at least one round of beneficiary verification. While the exact number of cards cancelled is not publicly disclosed in aggregate, state-level reports suggest tens of millions of cards were removed from rolls in states like Uttar Pradesh, Bihar, Madhya Pradesh, and Rajasthan during the digitisation process.
The flip side of this de-duplication is exclusion error. Some valid beneficiaries whose data was incorrectly matched or who had technical difficulties with Aadhaar seeding were excluded from the rolls. Field studies in several states found that Aadhaar authentication failures at ePOS machines sometimes left beneficiaries unable to draw their entitlement on the designated day.
Inclusion errors occur when non-poor households hold valid ration cards and draw PDS grain. Studies using National Sample Survey data have found inclusion error rates ranging from 20 to 25 percent, meaning roughly one in four beneficiary households in some states are not genuinely poor by common poverty metrics.
The problem is structural. State governments identify beneficiaries within central government quotas. Political incentives favour inclusion rather than exclusion. A local administrator who removes a household from the ration card list faces a political complaint. One who adds a non-poor household faces no immediate consequence. This asymmetry has historically produced slow expansion of beneficiary rolls and resistance to purging ineligible households.
The situation was further complicated by the PMGKAY extension. Since grain is now free, the marginal incentive to hold a ration card has increased even for households that are not poor. This raises the risk of inclusion errors expanding over time as the free food benefit becomes more valuable.
The statutory coverage under NFSA is fixed at 81.35 crore persons based on the 2011 Census and the 2011-12 household consumption expenditure survey. India’s decadal Census was due in 2021 but has not yet been conducted. As of June 2026, India is operating its largest social welfare programme on demographic data that is now 15 years old.
This creates two problems. First, population has grown. Approximately 79 lakh additional persons are estimated to be eligible for NFSA coverage within existing guidelines but have not been added because the statutory coverage number has not been updated. PRS India’s analysis of the 2026-27 Demand for Grants notes this gap explicitly. Second, poverty has declined. NITI Aayog’s Multidimensional Poverty Index shows the poverty ratio fell from 29.13 percent in 2013-14 to 11.28 percent in 2022-23. Many households that were poor when coverage was determined in 2013 are no longer poor. But they remain on the beneficiary rolls because there is no legal mechanism for automatic exclusion as households cross the poverty line.
The result is a system where the coverage is simultaneously too small (missing newly poor people and population growth) and too large (retaining non-poor households that graduated out of poverty). Both problems persist because no updated household consumption data exists to inform a rational revision of coverage.
The government’s technology-driven PDS reform programme has made measurable improvements. The One Nation One Ration Card scheme, operational across all states and UTs, allows portability of entitlements. Migrant workers who previously lost access to their PDS allocation when they moved to a different state can now draw grain at any FPS in the country. By 2025, over 97 percent of PDS transactions were being recorded biometrically or through Aadhaar-authenticated ePOS devices.
The SMART-PDS initiative, being rolled out in phases from December 2025, aims to further integrate beneficiary databases, grievance redressal, and supply chain monitoring. FCI’s over 2,500 godowns are linked to a national dashboard tracking stock positions. The Anna Mitra and Mera Ration apps allow beneficiaries to check their entitlements, locate nearby FPS outlets, and file complaints digitally.
The limits of the digital reform approach are that it addresses delivery efficiency but not targeting accuracy. A genuine non-poor household with a valid Aadhaar-linked ration card will be authenticated smoothly and will draw free grain every month. The problem is not delivery failure. The problem is that the eligibility criteria have not been revised using current data. Technology delivers entitlements accurately to whoever holds a card. But it cannot determine whether that card should have been issued in the first place.
Part VIThe Reform Debate
Four Reform Options Every Economist Has Proposed
The debate on reforming India’s food subsidy is not new. The 15th Finance Commission, the Economic Survey, the Commission for Agricultural Costs and Prices, and multiple planning documents have recommended some version of the same set of reforms. They have not been implemented. Understanding why requires understanding each option and the political economy that blocks it.
Direct Benefit Transfer (DBT) of Food Subsidy: Instead of giving grain in kind, the government would transfer cash directly to beneficiaries’ bank accounts equal to the market value of their grain entitlement. The beneficiary then buys food from any source they choose.
The case for it: DBT eliminates the need for the FCI’s vast procurement, storage, and distribution infrastructure. It reduces leakage because cash is harder to divert than grain. It gives beneficiaries choice, allowing them to buy the foods they prefer. It could reduce the government’s overall subsidy bill by eliminating FCI’s handling and distribution costs, which are embedded in the economic cost per kilogram.
The case against it: Cash transfers require functioning bank accounts, last-mile ATM infrastructure, and reliable mobile connectivity, which are still absent in parts of rural India. More fundamentally, cash given for food might be spent on other things. The in-kind grain transfer carries a political assurance of nutritional security that a cash transfer does not. Some economists also argue that removing the physical grain from the market could increase open market food prices, hurting the very poor who are not covered by NFSA. DBT has been piloted for food grain in Puducherry but has not been scaled nationally due to these concerns and political resistance from state governments that run their own PDS networks.
Increasing the Central Issue Price (CIP): Beneficiaries currently pay zero for their grain. The NFSA originally specified Rs 3 per kg for rice, Rs 2 for wheat, and Rs 1 for coarse grains. These prices were frozen in the Act for three years and were meant to be revised periodically. They were never revised upward. In 2023, they were revised downward to zero.
The case for it: Even a modest CIP of Rs 2 to 3 per kilogram, if applied to 81 crore beneficiaries receiving 5 kg per month, would generate Rs 9,720 crore to Rs 14,580 crore in annual cost recovery. More importantly, a non-zero price creates some self-targeting: households that are not genuinely poor have less incentive to maintain ration cards if they have to pay for grain. The NFSA itself contemplates periodic CIP revision linked to economic costs.
The case against it: No government has been willing to impose even a small charge for something that is now free and has been declared a legal right. The political optics of raising the price of rice from Rs 0 to Rs 2 per kilogram for 81 crore voters are difficult to manage. The BJP government, which extended the PMGKAY five years to 2028, has staked significant political capital on the free food message. No opposition party has proposed raising the CIP either.
Reducing Coverage Based on Updated Data: NITI Aayog’s MPI data shows that poverty fell from 29.13 percent in 2013-14 to 11.28 percent in 2022-23. If coverage were aligned with current poverty estimates rather than the 2011 baseline, the number of NFSA beneficiaries could theoretically be reduced from 81 crore to around 15 to 20 crore genuinely poor people. This would reduce the food subsidy bill dramatically.
The case for it: Concentrating the subsidy on the genuinely poor would allow a significantly higher transfer per beneficiary. Instead of spreading Rs 2.27 lakh crore across 81 crore people, the same amount distributed to 20 crore truly poor people would more than quadruple the per-capita transfer. This is the standard economic argument for targeted rather than broad-based welfare.
The case against it: The MPI poverty decline is contested. Some economists argue the methodology understates current poverty. More practically, removing 60 crore people from the free food rolls is politically unthinkable. Even if poverty has declined, food insecurity has not disappeared. Nutritional insecurity persists in households that are above the MPI poverty threshold. A household that crossed the poverty line in 2022 may be vulnerable again in 2025 due to job loss, illness, or a bad harvest. The practical mechanics of identifying who to remove, notifying them, and managing the political backlash make reduced coverage almost impossible to implement.
Reforming the MSP-FCI Procurement Nexus: The underlying driver of the rising food subsidy is the annual MSP increase. The government announces MSP for wheat and paddy every year. If the MSP rises 5 percent, the economic cost per kilogram rises by roughly the same amount, the subsidy per kilogram rises, and the total bill rises. Structural reform of the MSP-FCI system is the only way to contain costs at the source.
Reform proposals include: Limiting FCI’s role to procurement for buffer stock maintenance only, with distribution to states done through alternative channels. Introducing decentralised procurement where state agencies procure and distribute without routing through FCI, reducing handling costs. Rationalising the buffer stock norms, which currently result in far more grain being procured and stored than is needed. Introducing deficiency payments to farmers instead of physical procurement at MSP, where the government pays farmers the difference between MSP and market price without taking physical possession of the grain.
The political economy: MSP reform is the most politically sensitive of all food policy reforms. The 2020-21 farm laws, which attempted to create alternative marketing channels for farmers and reduce procurement dependence, were withdrawn by the government in November 2021 following sustained farmer protests, particularly in Punjab and Haryana. The withdrawal demonstrated that the political economy of MSP-linked procurement is too powerful for any government to confront frontally. FCI procurement at MSP provides income security to millions of farmers in wheat and rice-growing states, who are a key electoral constituency.
The Political Economy of Inertia
The food subsidy is large, rising, and structurally difficult to contain. Every reform proposal has a coherent economic rationale. None has been implemented at scale. This is not a failure of analysis. It is a failure of political will that reflects a genuine conflict between fiscal sustainability and electoral arithmetic.
India’s 81 crore PMGKAY beneficiaries represent approximately 59 percent of the total population and a far higher share of the voting-age electorate in rural constituencies where state assembly and Lok Sabha contests are won or lost. The free food programme is not a peripheral welfare benefit. It is a central element of the social contract between the Indian state and its poorer citizens, developed over seven decades and legally enshrined in the NFSA.
The question of whether taxpayers should bear this burden does not have a simple yes or no answer. Most taxpayers in India, particularly those in the lower middle class, also benefit from food security infrastructure in some form. The more precise question is whether the current design, coverage, delivery mechanism, and cost structure is optimal for delivering food security. On that question, the evidence strongly suggests room for improvement: better targeting, reduced leakage, eventual updating of beneficiary rolls to current demographic data, and a gradual return of the CIP to a non-zero level. These reforms are technically feasible. Whether they are politically feasible is a different question entirely.
Rs 2.27 Lakh Crore: What It Buys, What It Costs, and What Comes Next
India’s food subsidy machine is the largest food security programme in the world by beneficiary count. It feeds 81 crore people every month at zero cost to the recipient. It does this through a procurement, storage, and distribution infrastructure built over six decades and employing hundreds of thousands of people directly and indirectly. By the measures that matter most to its beneficiaries, it works. Most of the grain gets through. Food insecurity among the registered poor is lower today than at any point in India’s independent history.
But the fiscal cost is enormous and rising. Rs 2.27 lakh crore in FY27 is 5.2 percent of gross tax revenues. It is more than India spends on health or education from the Union Budget. The five-year PMGKAY extension is projected to cost Rs 12 lakh crore by 2028. The economic cost per kilogram continues to rise with each MSP announcement. The Central Issue Price is zero. The subsidy per kilogram equals the full economic cost. There is no automatic stabiliser.
The three structural problems are clear. Coverage is anchored to 2011 data in a country that was last censused in 2011 and is now 15 years older. The targeting errors, both inclusion and exclusion, have been reduced by technology but not eliminated. And the fundamental design, in which grain is procured at MSP, stored by FCI, and distributed free through 5.43 lakh fair price shops, was designed for an India of 1965. It has been reformed iteratively but not structurally.
The reforms the economics of the situation demand, revised coverage based on current poverty data, a non-zero CIP that recovers part of the cost, and a leaner FCI procurement role, are all politically difficult to near-impossible in the current environment. The PMGKAY extension to 2028 has locked the current architecture in for at least three more years. The Census, when it eventually happens, will force a reckoning with the 2011 beneficiary numbers. Until then, India’s taxpayers will continue to fund a free food programme that costs Rs 2.27 lakh crore a year, serves 81 crore people, and cannot be easily changed without an act of Parliament and the political will to spend the electoral capital that would require.
Frequently Asked Questions
The Union Budget for FY27 (2026-27) allocated Rs 2,27,429 crore for food subsidy. This represents 97 percent of the total allocation for the Department of Food and Public Distribution and approximately 0.50 percent of GDP. Almost all of this, Rs 2,27,429 crore, is allocated towards the Pradhan Mantri Garib Kalyan Anna Yojana (PMGKAY), under which 81.35 crore beneficiaries receive 5 kg of wheat or rice free of cost every month. This is broadly in line with the FY26 Revised Estimate of Rs 2.27 lakh crore (revised up 12.1 percent from the FY26 Budget Estimate of Rs 2.03 lakh crore due to higher grain procurement costs and OMSS expenditure).
Three factors drive the size and upward trend. First, coverage is large: 81.35 crore beneficiaries, based on the 2011 Census, is approximately 59 percent of India’s total population. Second, the Central Issue Price is zero since January 2023. The entire gap between the economic cost of procurement and distribution (Rs 41.73 per kg for rice and Rs 29.80 per kg for wheat in FY26) and the zero price charged to beneficiaries falls on the central budget. Third, the economic cost itself rises every year because the Minimum Support Price for wheat and paddy is revised annually, typically upward by 4 to 8 percent. Since the CIP cannot rise but the economic cost does, the subsidy per kilogram widens automatically each year. There is no cap, no index, and no automatic adjustment mechanism.
As of August 2025, 80.6 crore beneficiaries had been identified and were receiving free grain under PMGKAY and NFSA. The statutory maximum is 81.35 crore persons, which was fixed based on the 2011 Census under the National Food Security Act, 2013. This covers up to 75 percent of the rural population and up to 50 percent of the urban population. There is an estimated scope for an additional 79 lakh beneficiaries to be added within the existing statutory ceiling. The coverage ceiling has not changed since the NFSA was enacted in 2013. The decadal Census, due in 2021, has not yet been conducted, so the demographic basis for the coverage has not been updated for 15 years.
The Central Issue Price (CIP) is the rate at which the government issues grain from the central pool to beneficiaries through the PDS. Under the NFSA as originally enacted in 2013, the CIP was Rs 3 per kg for rice, Rs 2 per kg for wheat, and Rs 1 per kg for coarse grains. These prices were described in the NFSA as applicable for three years. They were never revised upward after that period, despite recommendations from multiple commissions and the RBI that they should track economic costs. In December 2022, the government decided to provide all NFSA grain free of cost from January 1, 2023. This brought the CIP to zero. The decision was subsequently embedded in the PMGKAY extension for five years from January 2024 to December 2028. Reintroducing a non-zero CIP now would require amending the scheme notification, which is administratively feasible but politically very difficult.
The Food Corporation of India (FCI) is a statutory body under the Food Corporations Act, 1964. It was established on January 1, 1965. FCI handles over 70 percent of India’s food subsidy expenditure. Its core role is to procure wheat and paddy from farmers at the government-declared Minimum Support Price, store the grain at its network of over 2,500 godowns (total storage capacity: 482 lakh metric tonnes as of November 2025), and transport the grain in bulk to state distribution points. State governments and their agencies then distribute the grain to Fair Price Shops, from which beneficiaries draw their entitlement. The food subsidy essentially reimburses FCI for the difference between the economic cost of all these operations and the zero revenue it receives from beneficiaries. In FY26, FCI’s own revised expenditure estimate for the year was Rs 1.7 lakh crore, up from an initial estimate of Rs 1.4 lakh crore, reflecting higher procurement volumes and rising MSP costs.
The main proposals that have been extensively discussed are: Direct Benefit Transfer of a cash equivalent instead of in-kind grain distribution; reintroduction of a non-zero Central Issue Price to recover part of the cost; reduction in coverage to align with current poverty data rather than 2011 Census projections; and reform of the MSP-FCI procurement system to reduce the underlying cost of grain before it enters the distribution chain. Each of these has been recommended by the 15th Finance Commission, multiple Economic Surveys, and expert committees. None has been implemented at scale. The reasons differ by proposal. DBT faces infrastructure and political resistance. CIP restoration would mean raising the price of something now free for 81 crore voters. Coverage reduction would mean removing people from free food rolls in a country where hunger and nutritional insecurity are still visible. MSP reform proved politically explosive in 2020-21 when the farm laws were withdrawn after sustained protests. The PMGKAY extension to December 2028 effectively locked the current no-CIP, maximum-coverage, in-kind-distribution architecture in place for at least five years from January 2024.
Disclaimer: This article is for informational and educational purposes only and is current as of June 2026. All figures are drawn from Union Budget documents, the Department of Food and Public Distribution’s Demand for Grants, PRS India budget analyses, FCI official data, NITI Aayog publications, and government press releases. This article does not constitute policy advice or an endorsement of any particular reform option. Nothing herein constitutes investment advice.
