Beyond the Finance Commission: CSS, discretionary transfers and special category status

Detail

1. Two routes for Centre-to-state money

The Centre sends money to states by two routes.

• Statutory transfers are required by the Constitution. Their size and sharing follow the advice of the Finance Commission (Art. 280).
• Discretionary transfers are chosen by the Centre. They flow through ministries and schemes under Art. 282, which lets the Union give grants "for any public purpose".

Statutory transfers	Discretionary transfers
Flow on Finance Commission advice	Flow through ministries and schemes at the Centre's choice (Art. 282)
Tax devolution (Art. 270: states get a share of the divisible pool of central taxes) + Art. 275 grants-in-aid (fixed grants to states that need help)	Centrally sponsored schemes (CSS), central sector schemes, special assistance, NDRF (National Disaster Response Fund) releases
Formula-based. Untied or only lightly tied	Conditional. The Centre sets the design

• Why the difference matters:
• Statutory money comes by formula. States can plan with it and spend it on their own priorities.

• Discretionary money comes with conditions. The Centre decides what it is spent on, and how and when.

• Current Finance Commission frame: the 16th Finance Commission (2026-31) kept the states' share of the divisible pool at 41%, the same as the 15th FC [12].

• It discontinued revenue deficit grants, sector-specific grants and state-specific grants [12].

• With less untied grant money, states now depend more on discretionary channels such as CSS and SASCI.

• Earlier jump: the 14th FC (2015-20) raised the states' share from 32% (13th FC) to 42% [10].

2. Special Assistance to States for Capital Investment (SASCI)

• What it is: the Centre gives states 50-year interest-free loans to spend on capital expenditure, meaning spending that builds lasting assets such as roads, bridges and buildings. It started in 2020-21.
• Reform link: part of each year's amount is released only when states carry out reforms.

• Example: in the power sector, SASCI money can be used to repay debt after a state privatises its DISCOMs (electricity distribution companies) [12].

• Growth in size:

• About ₹12,000 crore (2020-21), rising to ₹1,50,000 crore (2025-26 BE) [3].
• The Union Budget 2025-26 proposed ₹1.5 lakh crore for these loans and for reform incentives [2].

• States received ₹4,01,276 crore in total between 2020-21 and 2025-26 (till 11 August 2025) [4].

• Dependence: states were estimated to fund 19% of their capital outlay through SASCI loans in 2024-25, up from 2.9% in 2020-21 [3].

• Worked example: a state gets ₹1,000 crore under SASCI.
• If it had borrowed the same amount from the market at an assumed 7%, it would pay ₹70 crore interest every year.
• Under SASCI it pays ₹0 interest and repays the principal after 50 years.

• In practice, the loan works almost like a grant.

• Legal angle: SASCI is a loan. A state that owes money to the Centre needs the Centre's consent to borrow more (Art. 293(3)). This gives the Centre extra influence over the state.

3. The planning-era channel (now closed)

• Normal Central Assistance (NCA): Planning Commission money given to support state five-year plans. States shared it by formula.
• Gadgil formula (1969): the first formula for dividing NCA. It was adopted by the National Development Council (NDC), which the Prime Minister chaired and whose members included Union Ministers, all Chief Ministers and Planning Commission members [7].
• Gadgil-Mukherjee formula (1991): the revised formula.
• Step 1: 30% of NCA was set aside for special category states. The other 70% went to the remaining states [7].
• Step 2: the remaining states shared their pool by these weights [7]:
  • Population: 60%
  • Per capita income: 25%
  • Fiscal performance: 7.5%
  • Special problems: 7.5%
• Grant-loan mix: special category states got NCA as 90% grant : 10% loan. Other states got 30% grant : 70% loan [7].

• Worked example: suppose NCA is ₹10,000 crore.

  • ₹3,000 crore goes to special category states, and ₹2,700 crore of that is a grant.
  • ₹7,000 crore goes to the other states, and only ₹2,100 crore of that is a grant.

• End of the channel:

• The Planning Commission was abolished (2015) and replaced by NITI Aayog. NITI Aayog is a think-tank and does not allocate funds.
• The plan/non-plan distinction was merged (2017-18). The Budget now classifies spending as revenue vs capital instead.
• NCERT Class 12, Government Budget and the Economy, still describes plan and non-plan expenditure. That classification is now outdated.

4. Centrally sponsored schemes (CSS)

• Centrally sponsored scheme (CSS): a scheme designed by the Centre, implemented by states and funded jointly by the Centre and the state. Examples: MGNREGA and PMGSY.
• Central sector scheme: a scheme fully funded and run by the Centre on subjects in the Union List. The state pays no share.
• NITI Aayog Sub-Group of Chief Ministers (2015) grouped CSS into three kinds:
• Core of the core: social-protection umbrella schemes and MGNREGA. These keep their existing funding patterns.
• Core schemes: 60:40 (Centre:state) for general states. 90:10 for NE and Himalayan states [5].

• Optional schemes: states choose whether to join. 50:50 for general states. 80:20 for NE and Himalayan states [5].

• Special patterns:

• 90:10 for NE and Himalayan states.
• 100% central for UTs without a legislature.

• Some schemes are fully central in every district, for example Beti Bachao Beti Padhao [13].

• Worked example: take a core scheme with a ₹1,000 crore state outlay.

State type	Centre pays	State pays
General state (60:40)	₹600 crore	₹400 crore
NE/Himalayan state (90:10)	₹900 crore	₹100 crore
UT without legislature	₹1,000 crore	₹0

• Appraisal cycle: every five years, the Centre appraises and re-approves all CSS and central sector schemes. The cycle is aligned with the Finance Commission period [11].

5. Fund-flow controls: SNA and SNA-SPARSH

• Single Nodal Agency (SNA) system (2021): each state names one agency per CSS. All central and state money for that scheme sits in one bank account, tracked on PFMS (Public Financial Management System).

• Why it was needed: unspent money used to lie idle in many different accounts.

• SNA-SPARSH: a newer model for "just-in-time" release. Money leaves the Treasury only when an actual payment is due, not in advance [6].

• FY 2025-26 releases used the SNA module till 30 June 2025 and SNA-SPARSH from 1 July 2025 [6].
• Effect on the Centre: less idle cash sits with states, so the Centre borrows less.
• Effect on states: they lose the flexibility of holding float money, and they depend on the Centre's release timing.

6. Critiques of CSS

• Less state autonomy:
• The Centre sets conditions and uses one-size-fits-all designs.

• Kerala and Bihar have very different needs, but they get the same scheme template.

• Matching shares crowd out state priorities:

• To receive Centre money, a state must put in its own 40%.
• That 40% comes from the state's untied money.

• So less money is left for the state's own schemes.

• Tighter control over money flow: SNA and just-in-time release increase central control over how and when money moves.

• Weakening of the FC channel: when discretionary transfers grow faster than FC transfers, more of the states' money comes with conditions attached.

7. Special category status (SCS)

• Special category status: a label that gave some states more favourable central assistance. The National Development Council created it in 1969, together with the Gadgil formula.
• Reason: some states have a low resource base because of their geography and history. They cannot raise enough money for development on their own [7].
• Criteria:
• Hilly or difficult terrain.
• Low population density or a large tribal share.
• Strategic location on international borders.
• Economic and infrastructure backwardness.

• Non-viable state finances.

• Who got it:

• First 3 states: Assam, Nagaland and Jammu & Kashmir [8].
• 8 added later: Arunachal Pradesh, Himachal Pradesh, Manipur, Meghalaya, Mizoram, Sikkim, Tripura and Uttarakhand [7][8].

• Total: 11 states. These are the 8 NE states (including Sikkim), Himachal Pradesh, Uttarakhand and the erstwhile state of J&K.

• Benefits:

• 30% of NCA, as a 90:10 grant-loan mix [7].
• 90:10 CSS funding.

• Tax concessions, for example on excise duty, to attract industry.

• Finance Commission position:

• SCS never came from any Finance Commission. It came from an NDC resolution [9].
• It is not part of the FC's Terms of Reference, as clarified by the 15th FC Chairman N. K. Singh [9].

• The FC does not distinguish between special and general category states [7][9].

• 14th FC: made no special/general category distinction in devolution. It covered these states' needs through the higher 42% share and revenue deficit grants [10].

• After 2015: with the NDC and Planning Commission inactive, SCS survives mainly through the 90:10 CSS pattern for NE and Himalayan states.
• Continuing demands:
• Andhra Pradesh (a promise made at the 2014 bifurcation).
• Bihar.
• Odisha.

8. Raghuram Rajan Committee (2013): an index instead of a label

• Proposal: a composite development index that ranks states by how underdeveloped they are. It was meant to replace the yes/no special category label.
• Remit: the committee was asked to suggest ways to identify backwardness of states and to show how it should shape the devolution of central funds [11].
• Classification by index score [11]:
• ≥ 0.6 → "Least Developed"
• 0.4 to 0.6 → "Less Developed"

• < 0.4 → "Relatively Developed"

• "Least Developed" states (10) [11]:

• Arunachal Pradesh, Assam, Bihar, Chhattisgarh, Jharkhand, Madhya Pradesh, Meghalaya, Odisha, Rajasthan and Uttar Pradesh.

• Note: Bihar and Odisha, which demand SCS, qualify as "Least Developed" under this index.

• Idea behind it: need is a continuous scale. Funds should rise smoothly with need, not jump when a state gets a label.

Prelims Hooks

• Art. 282 is the constitutional basis for discretionary grants (CSS and central sector schemes). Art. 275 provides statutory grants-in-aid on FC advice.
• Special category status came from the National Development Council (1969), not from any Finance Commission. It is not in the FC's Terms of Reference.
• First three SCS states: Assam, Nagaland and J&K. The total later reached 11.
• Gadgil-Mukherjee formula (1991) weights: population 60%, per capita income 25%, fiscal performance 7.5%, special problems 7.5%. 30% of NCA went to special category states.
• SCS grant-loan mix: 90:10. General states: 30:70.
• CSS funding patterns: core 60:40 (general) and 90:10 (NE/Himalayan). Optional 50:50 and 80:20. UTs without a legislature: 100% central.
• SASCI: 50-year interest-free loans for capital expenditure, since 2020-21. ₹1.5 lakh crore in Budget 2025-26.
• Rajan Committee (2013): composite index score ≥ 0.6 = Least Developed.
• 16th FC (2026-31): states' share stays at 41%. Revenue deficit grants discontinued.
• Trap: NITI Aayog does not allocate plan funds. Plan/non-plan was merged in 2017-18, and revenue/capital is the current classification.

Mains Points

• Rising discretionary transfers weaken fiscal federalism.
• CSS, SASCI and reform-linked loans come with conditions and grow faster than formula-based FC transfers.
• This pulls policy choice towards the Centre.

• Counter-view: conditions push reforms and ensure national minimum standards in health, education and rural jobs.

• SASCI is a trade-off.

• It has lifted state capital spending (19% of state capital outlay in 2024-25).

• But it builds dependence, links money to the Centre's reform choices, and adds to debt owed to the Centre (Art. 293).

• Rationalise CSS.

• Have fewer and larger schemes.
• Give states flexible funds within each scheme.
• Keep matching shares low for poorer states.

• Balance SNA-SPARSH efficiency against states' need for cash flexibility.

• SCS vs an index-based approach.

• Demands from Andhra Pradesh, Bihar and Odisha show that a yes/no label causes political bargaining.
• An index like the Rajan Committee's links transfers to measured need in a transparent way.
• After the 14th FC, need is better handled through the devolution formula than through a label.