India has mastered the politics of giving. It has yet to build a system for stopping, redesigning and reallocating.
India’s welfare state has acquired an impressive piece of plumbing. Direct Benefit Transfer has channelled almost Rs 53 lakh crore through 320 schemes across 56 ministries, with estimated savings of more than Rs 5.14 lakh crore. What once depended on layers of intermediaries can now be transferred directly to beneficiaries and tracked at scale. The harder problem is no longer delivery. It is exit. A scheme may outlive the problem it was created to solve, become poorly designed or deliver less value than an alternative use of the same money. Yet once it acquires beneficiaries and political expectations, withdrawal becomes harder than launch. The answer is to make continuation conditional: renew what works, redesign what does not and close what has run its course.
The State Learned To deliver
The old problem in Indian welfare was whether money reached the intended beneficiary. Aadhaar, bank accounts and digital payments have made that problem much smaller. The harder question is whether the intervention improved the outcome it was designed to change. Every programme faces three tests: was the money spent, did it reach the intended beneficiary, and did it improve the outcome it promised? Digital systems increasingly answer the first two. The third requires evidence that a payment changed behaviour, raised incomes, improved learning or delivered some other measurable gain. Counting a payment is easy. Proving that it changed a life is the harder measure of government.
The scale of the commitment makes this important. The Sixteenth Finance Commission estimates that subsidies and transfers across 21 states rose from Rs 3.86 lakh crore in 201819 to Rs 9.73 lakh crore in 2025-26, taking their share of revenue expenditure from 15.6% to 19.8%. Unconditional cash transfers alone now account for Rs 1.96 lakh crore. The issue is not welfare spending itself, but whether each programme remains the best use of the next rupee. A programme costing Rs 10,000 crore a year becomes a Rs 1 lakh crore commitment over a decade, yet budgets are voted one year at a time. The benefit is visible today; the opportunity cost is dispersed across future budgets. This is how a temporary intervention becomes a permanent commitment. A scheme gathers beneficiaries, builds an administrative constituency and becomes politically harder to touch. Eventually, its existence becomes its strongest argument for continuing.
Counting Payments is not measuring results
Digital delivery can create an illusion of success. A ministry can report millions of transfers completed and beneficiaries covered without knowing whether the underlying problem has improved. That changes what government should measure. Programmes should be judged against outcomes— higher earnings, better learning, improved health or greater productivity— not simply expenditure and coverage. A department judged by money spent has an incentive to spend; one judged by results has an incentive to ask whether its intervention still works.
Every Benefit Creates a Constituency
Policy persistence is not simply an administrative accident. A continuing benefit has an obvious political advantage: its beneficiary can see the money, associate it with the government and expect it again. A Rs 1,000 monthly transfer may be modest in a state budget but highly visible to the household receiving it; a bridge built with the same money is harder to attribute to a particular government. Adding a benefit creates a constituency; removing it creates an opponent. A government can announce a scheme today while leaving its eventual cost to a future government, which inherits the beneficiaries but not the political credit. The political system rewards the announcement of benefits more reliably than their withdrawal.
Bureaucracy has reasons to keep the machine running
The same incentive operates inside government. A department administering ten schemes has more files, reporting requirements and reasons to defend its staff and budget. Consolidating them may simplify life for citizens while shrinking the institution responsible for administering them. One scheme creates reporting requirements; reporting creates staff; staff create institutional ownership; ownership creates another reason to preserve the scheme. This is not necessarily corruption. It is institutional self-preservation. If an evaluation produces another report but no budgetary consequence, failure carries little institutional cost.
Recurring spending crowds out the future
The fiscal cost eventually becomes harder to hide. Recurring transfers compete with capital spending because states have limited fiscal room. When salaries, pensions, interest payments, subsidies and transfers consume more revenue, raising capital expenditure requires stronger tax collections, more borrowing or cuts elsewhere. The trade-off matters because capital spending has a different economic life. A transfer supports consumption today; a road, irrigation system or power network can raise productivity for years. The danger is not welfare itself, but recurring commitments becoming hardest to reduce when fiscal space is most valuable. A permanent welfare commitment can leave less room for the public investment that could raise incomes and reduce the need for transfers.
Once a Benefit becomes a right
Stopping a scheme becomes harder when its benefit has become a legal right. The National Food Security Act creates entitlements to subsidised food grains for eligible households, alongside nutritional entitlements for specified groups. Once a benefit is embedded in primary legislation, a ministry cannot simply withdraw it newal date. The test should be simple: Has the original problem changed? Is the programme still producing the intended outcome? Is this still the best use of the money? through an administrative order. Parliament has to amend the law. The state can sunset a scheme; it cannot casually sunset a right. Fiscal federalism adds another constraint. The Rs 9.73 lakh crore state figure conceals large differences in fiscal capacity. A state with strong own-tax revenues can withdraw one programme and redirect the money; a state with little fiscal room may have to cut another commitment. A uniform sunset timetable would therefore impose the greatest adjustment on states least able to absorb it. The Centre may design the exit, but states often carry the political and fiscal cost.
Make every rupee earn its renewal
Brazil offers a useful comparison. Bolsa Família was replaced by Auxílio Brasil in 2021 and redesigned again as Bolsa Família in 2023. The lesson is simple: protecting a programme’s purpose does not require protecting its architecture. India should apply the same principle. A subsidy may need to become targeted. A universal benefit may need to become conditional. A crisis programme may need to shrink when the crisis passes. The Sixteenth Finance Commission has recommended sunset or exit clauses for subsidies and transfers, while noting that very few major schemes have actually been phased out. Every major scheme should have a baseline, measurable outcomes, independent evaluation and a reIf yes, renew it. If partly, redesign it. If no, close it.
There is also a case for simplifying income support. If several programmes address insufficient household income, a negative income tax could provide transfers below a defined threshold, tapering as earnings rise. Its attraction is consolidation: income becomes the organising principle rather than an expanding catalogue of schemes. This would not replace food security, healthcare, education, pensions or disability support. It would make income support simpler to administer and easier to review. The broader principle matters more than the instrument: fewer schemes, clearer objectives and a system capable of changing when circumstances change.
A new programme creates a headline. An old programme creates beneficiaries who know exactly what they might lose. The minister gets credit for the first and a delegation for the second. India has mastered the politics of giving. The harder reform is to make every rupee compete again for its place in the budget.
*Prof. Vikas Singh is a Professor at IIM Nagpur and Visiting Faculty at the Indian School of Business (ISB), and writes on business, policy and India’s economic transformation.