The states that spend most on welfare have the least to show for it
India's biggest social-sector spenders record some of its worst welfare outcomes — while the states that spend the least on welfare, as a share of their budgets, are the ones pulling ahead. A look at what the numbers behind state budgets actually say, and what they can't.
Few budget lines say more about a government's priorities than what it spends on schools, hospitals, nutrition, and welfare. Across the last decade, the "social sector" has become the single largest component of state spending in India. But averages hide a striking divergence: some states devote more than half of their budgets to social sectors, while others allocate barely a fifth — and the two groups are not who you would guess.
The states that budget the largest share for welfare — Jharkhand, West Bengal, Bihar, Chhattisgarh — also carry the country's worst maternal mortality, anaemia, and multidimensional-poverty numbers. The states near the bottom of the spending table — Kerala, Tamil Nadu — post the best outcomes in the country. High spending is tracking poor outcomes, not curing them.
Who spends the most on welfare?
In 2025-26, Jharkhand, West Bengal, Delhi, Chhattisgarh, and Bihar top the table, each directing roughly half of total budgetary spending to the social sector. Jharkhand leads at 54.8% — up from 49.9% a decade earlier. West Bengal's share climbed even faster, from 47.4% to 54.2%. At the other end, Punjab commits just 21.3%, with Kerala and Tamil Nadu also in the bottom five despite their long-standing reputation for strong human development.
View data table
| State | Social-sector share (2025-26) |
|---|---|
| Jharkhand | 54.8% |
| West Bengal | 54.2% |
| Delhi | 54.1% |
| Chhattisgarh | 50.3% |
| Bihar | 49.8% |
| All-state average | 42.2% |
| Tamil Nadu | 33.6% |
| Kerala | 32.2% |
| Punjab | 21.3% |
Read on its own, this table looks like a straightforward story of generosity: poorer, more deprived states spending hard to catch up. And in part it is. Jharkhand, Bihar, and Chhattisgarh are among India's poorest and most populous states; they carry enormous unmet needs in health, education, and nutrition, so a large social-sector share is exactly what you would expect. The puzzle appears only when you put spending next to results.
More money, worse results
Take maternal mortality — one of the sharpest tests of whether a health system reaches poor women. Chhattisgarh, which spends 50.3% of its budget on social sectors, records 146 maternal deaths per 100,000 live births — the highest among the states analysed. West Bengal, another heavy spender at 54.2%, records 104, more than triple Kerala's rate. Kerala and Tamil Nadu, both in the bottom five for social-sector share, record 30 and 35 respectively.
View data table
| State | Social-sector share | Maternal deaths / 100k |
|---|---|---|
| Kerala | 32.2% | 30 |
| Tamil Nadu | 33.6% | 35 |
| West Bengal | 54.2% | 104 |
| Chhattisgarh | 50.3% | 146 |
The same pattern repeats across other measures. Bihar, a big spender, has the highest poverty headcount ratio (33.8%) and the highest multidimensional-poverty index of the group; nearly two-thirds of its pregnant women are anaemic. In Jharkhand, 28.8% of the population remains multidimensionally poor and anaemia affects more than half of expectant mothers. Kerala, by contrast, has almost eliminated multidimensional poverty, and its anaemia rate among pregnant women (31.4%) is less than half the national average.
The states doing well today are not spending less because welfare doesn't work. They are spending a smaller share because decades of earlier investment already built the schools, clinics, and public systems that hold their outcomes up. Kerala and Tamil Nadu are living off a stock of past spending; Bihar and Jharkhand are still paying to build the stock. Cross-sectional budget shares capture the flow, not the accumulated stock — which is exactly why the snapshot inverts.
Where the gender rupee goes
A second layer of the story is how much of the budget is directed at women specifically. Here the ranking scrambles again. Jharkhand earmarks 35.5% of its budget for women-focused schemes — the highest of any state — but so does Tamil Nadu, at 29.8%, second only to Jharkhand and far above its overall social-sector rank would suggest. Both sit well above the national average of 9.4%. Delhi, a top-five social-sector spender overall, directs just 9.5% to women — and records the lowest female labour-force participation in the country, at 10.4%.
View data table
| State | Women-focused share of budget |
|---|---|
| Jharkhand | 35.5% |
| Tamil Nadu | 29.8% |
| West Bengal | 26.3% |
| Kerala | 21.4% |
| Delhi | 9.5% |
| National average | 9.4% |
That Tamil Nadu can post both a low overall social-sector share and one of the highest gender-budget shares is a useful reminder that these numbers measure allocation, not effort or effectiveness. A state that has already achieved near-universal schooling can shift its marginal rupee toward targeted women's schemes; a state still building basic coverage cannot.
What this data can and can't tell you
It is tempting to turn a chart like the maternal-mortality scatter into a slogan — "welfare spending doesn't work." That would be a misreading, and the reason it would be a misreading is worth stating plainly, because it is the whole point of a data dive.
- Budget shares are not spending levels. A "share of total budget" says nothing about rupees per person. A poor state spending 54% of a small budget may still spend less per capita than a rich state spending 32% of a large one.
- Allocation is not delivery. Money budgeted is not money spent well. Leakage, absorption capacity, and administrative quality sit between the budget line and the clinic — and they vary enormously across states.
- Stocks vs flows. Today's outcomes reflect decades of past investment (the stock), while budget shares capture this year's spending (the flow). Kerala's advantage was bought in the 1970s–90s; it shows up now as a licence to spend a smaller share.
- Association is not causation. The spend-vs-mortality pattern is a correlation across a handful of states, driven by the underlying fact that poor states have both high need and high spending. It does not mean spending causes worse outcomes.
- Coverage is partial. The RBI analysis excludes north-eastern states and union territories other than Delhi, and works from budget estimates (BE), which are intentions, not audited actuals.
Read this way, the data delivers a sharper and more useful message than the slogan: where a state is on its development journey shapes how much it needs to spend — and spending share is a poor scoreboard for judging whether that money is working. To judge that, you need per-capita spending, outcome trends over time, and delivery data — the natural subjects of the next data dives.
- Reserve Bank of India, State Finances: A Study of Budgets of 2025-26 — social-sector expenditure shares, gender-budget statements, and outcome indicators.
- Shikha Chaturvedi, "From Bihar to Kerala, state Budgets chart divergent paths in social welfare," Business Standard, 2026 — reporting and analysis that surfaced these figures.
- PRS Legislative Research, State Budget analyses — for readers who want to trace individual state budgets.
ImpactMojo Data (2026). "The states that spend most on welfare have the least to show for it." ImpactMojo Data Dives. Retrieved from https://impactmojo.in/DataDives/state-welfare-budgets.html
Have a dataset worth digging into?
Data Dives are independent investigations built from public data. If you know a number that deserves a closer look, tell us.
Pitch a Data Dive →