India's Corporate Average Fuel Efficiency norms set a mass-based fleet CO₂ target for every passenger-vehicle manufacturer. The Ministry of Power's official CAFE 2027 draft — published 16 July 2026, effective 1 April 2027 — adds a codified escalating buyout, bilateral credit trading and a split 3 + 2 compliance block, turning fleet planning into a traded position.
CAFE is not a cap-and-trade market. Each manufacturer is judged on the sales-weighted average CO₂ of everything it sells in a reporting year, against a target derived from the kerb mass of that same fleet. Sell heavier vehicles and the target loosens; sell lighter ones and it tightens. Beat the target and the manufacturer accrues credits, miss it and it accrues debits.
Three institutions split the work. BEE, under the Ministry of Power, designs the scheme and sets the target curve. MoRTH enforces it. ARAI tests and certifies the per-model CO₂ values that feed the fleet average. The legal foundation and the penalty schedule sit in the Energy Conservation (Amendment) Act, 2022.
What changed on 16 July 2026 is the market layer. The Ministry of Power's official draft notification, F. No. 10/3/2021-EC, replaced the consultation-stage figures that circulated through 2025 and early 2026 with a numbered document containing three genuinely new provisions: an escalating BEE buyout, bilateral credit trading in place of EU-style pooling, and a split 3 + 2 compliance block.
The draft was opened for objections for 21 days, so provisions may change before finalisation. The mass-based target formula and the year-by-year curve carry over unchanged from the earlier 25 September 2025 draft.
Every manufacturer's target is a function of its sales-weighted kerb mass, pivoting on a reference mass. The official draft moved that pivot to 1,229 kg, with a reference target of about 78.9 g CO₂/km across FY28–FY32 — down from the CAFE-II baseline of roughly 113 g CO₂/km that runs through FY27.
The target is set by the fleet's sales-weighted kerb mass against a 1,229 kg pivot. A heavier fleet earns a looser target, a lighter fleet a tighter one — which is why the mass-curve slope is the central fault-line between small-car and SUV-heavy manufacturers.
The reference target of about 78.9 g CO₂/km applies across FY28–FY32, but compliance is assessed in two blocks — three years to FY30, then two to FY32 — not year by year.
The draft applies to the M1 category — passenger vehicles carrying up to eight people besides the driver. Small-volume manufacturers are exempt from the penalty provisions.
Targets stay on the MIDC cycle. Dual MIDC and WLTP declaration began 1 April 2026, and ARAI published AIS:175 in March 2026 — but the CAFE-specific WLTP switch needs a MoRTH notification and a MoP conversion factor, neither of which has been issued.
Penalty is computed after credits are settled, as total debit divided by block sales multiplied by the 23.7135 petrol-equivalent factor, under Section 26 read with 27–28. Proceeds split 90/10 between the States and the Centre.
In March 2026 the regulator levied ₹2,728 Cr in CAFE-II penalties across 9 OEMs for FY23–25. Climate Decode's five-year model brackets industry-wide CAFE 2027 exposure at roughly ₹8,000–15,000 Cr for FY28–32.
The official draft turns CAFE from an annual filing exercise into a traded position. A surplus manufacturer's credit is a genuinely saleable asset; a debit has a known, rising price. Timing is now itself a cost.
A manufacturer can offset a debit by buying credits from BEE at a price that rises every year. A gram left unsettled costs 80% more in FY32 than in FY28 — the rising price deliberately rewards trading and banking early.
Low-carbon powertrains count for more than one vehicle in the fleet average. Effective volume is Nᵢ = vᵢ × nᵢ, so the multiplier scales the whole sales line, not just the emissions figure.
“Pooling” in the CAFE draft means the bilateral exchange or trade of credits on mutually agreed terms, with the result furnished to the designated agency. There is no joint fleet assessed as one manufacturer, no nominated pool manager and no cap on counterparties. The earlier up-to-three-OEM pool is gone.
Credits and debits sit in a passbook and carry forward within the compliance block. Any credit left unsettled at block-end lapses — so surplus has a shelf life, and the block boundary is a hard planning date.
Three years from FY2027-28 (FY28–30), then two from FY2030-31 (FY31–32). Penalty is assessed at the end of each block, on the entire block sales, after credits are settled — not annually.
Off-cycle technologies are capped at 9 g/km in total, now awarded through a formal 12-technology menu at 1 g each (Table 1.3) — replacing the open-ended cap with a defined list.
Now explicit: 8% for E20-plus petrol including SHEV and PHEV, 22.3% for flex-fuel ethanol and flex-SHEV, 5% or the notified CBG share for CNG, and diesel per MoPNG.
Across the 15 passenger-vehicle OEMs, position is structural. Maruti, at roughly 1.87M units in FY26, carries the lightest fleet and therefore the tightest mass-curve target. Tata, at about 14% BEV, is positioned as a structural credit seller.
Where your fleet stands under India CAFE 2027, what the block costs, and which levers close the gap — mapped by the team.
TerraNova for CAFE, a joint solution with Meta Materials Circular Markets, turns the annual filing exercise into a live operating picture — built for the regulatory, product-planning, treasury and investor-relations desks.
A variant-level fleet ledger that stays current, with CNF, cᵢ and vᵢ auto-computed from your last-reported iCAT sheet. Generates the AIS-137 Appendices in official format and establishes your position today — fed by VAHAN, SIAM and ARAI data.
Applies the mass-based target formula against your sales-weighted fleet, layering Table 3 super-credits, the off-cycle derogation and the Carbon Neutrality Factor — so the gap you plan against is the gap the regulator will compute.
Every decarbonisation lever ranked by cost per gram closed — off-cycle technologies, mass reduction, powertrain shifts, strong hybrids and BEV ramp. Launch and switch scenarios in one click, and the ranking re-sorts when a draft provision moves.
Models the capex offset across PLI-Auto, PLI-ACC, PM E-DRIVE, state EV policies and GST alongside the CAFE super-credit value — so the per-BEV value stack is priced, not assumed.
Five-year forecasting across the 3 + 2 block, BAU against with-plan, with the escalating buyout schedule and Section 26 penalty math built in. Answers bank, trade or buy out — and when, given a price that rises every year.
An in-house demand-and-supply model projects each of the 15 OEMs long or short, so you can identify trading counterparties and cover early. Trade tape, BEE price band and the policy news feed in one screen.
CAFE 2027 figures on this page come from draft notification F. No. 10/3/2021-EC, dated 16 July 2026, which was open for objections for 21 days. Provisions may change before the rules are finalised. Where a value carries over from the earlier 25 September 2025 consultation draft — notably the mass-based formula and the year-by-year curve — that is noted. Exposure brackets attributed to Climate Decode are modelled, not regulatory figures.
A seven-part series on the CAFE regime — from first principles through the official 2027 draft to the credit market, modelled in full.
The 16 July 2026 draft — a 1,229 kg mass pivot, an escalating BEE buyout, uncapped bilateral trading and a 3 + 2 block.
The rulebook from first principles — the BEE/MoRTH split, the legal foundation in the EC Act, and the 113 g/km CAFE-II baseline.
Who is long, who is short, and where the clearing price lands across FY28–FY32 under four scenarios.
Answers on scope, targets, the buyout, credit trading and what the 16 July 2026 draft actually changed.
CAFE 2027 is the third phase of India's Corporate Average Fuel Efficiency norms for M1 category passenger vehicles, drafted to take effect on 1 April 2027. It sets a fleet-average CO2 target for each manufacturer based on the kerb mass of the vehicles it sells. Manufacturers below their target earn credits; those above accrue debits that must be settled by trading, banking or a buyout paid to BEE.
The Ministry of Power published the official draft notification, F. No. 10/3/2021-EC, on 16 July 2026, with effect from 1 April 2027. It was opened for objections for 21 days, so provisions may still change before the rules are finalised. The mass-based target formula and year-by-year curve carry over from the earlier 25 September 2025 consultation draft.
The CAFE 2027 draft codifies an escalating BEE buyout, priced per gram of CO2 per km: 2,500 rupees in FY2028, rising through 3,000, 3,500 and 4,000 to 4,500 rupees by FY2032. A gram left unsettled in FY32 therefore costs 80 percent more than the same gram in FY28, which deliberately rewards trading and banking early over deferring to a late buyout.
No. In the CAFE 2027 draft, pooling means the bilateral exchange or trade of credits between manufacturers on mutually agreed terms, with the result furnished to the designated agency. There is no joint fleet assessed as a single manufacturer, no nominated pool manager and no cap on counterparties. The earlier construct of up to three manufacturers per pool is gone.
Super-credits are confirmed and extended: 3.0 for battery and range-extended electric vehicles, 2.5 for plug-in hybrids and flex-fuel strong hybrids, 1.6 for strong hybrids and 1.1 for flex-fuel ethanol. Off-cycle technologies are capped at 9 g/km in total, awarded through a formal 12-technology menu at 1 g each. A Carbon Neutrality Factor applies on top, at 8 percent for E20-plus petrol and 22.3 percent for flex-fuel ethanol.
TerraNova for CAFE turns the annual filing exercise into a live operating picture: a variant-level fleet ledger that generates the AIS-137 Appendices, MAC-ranked decarbonisation levers, a five-year forecast across the 3+2 block, and market intelligence on which manufacturers are long or short. Climate Decode also publishes a seven-part India CAFE series covering the regime end to end.
See how TerraNova for CAFE turns the fleet ledger into a BEE-ready reporting pack, a MAC-ranked plan, and a five-year block forecast with the buyout schedule built in.