Across the world, states are turning to greener technologies and alternative fuels in fulfilment of their environmental and developmental commitments. Those commitments are not a single, harmonious command. Sustainable Development Goal 7 seeks affordable and clean energy, while Goal 10 seeks reduced inequalities, and the two can pull against each other. A green measure whose costs fall hardest on those least able to bear them advances clean energy even as it quietly deepens inequality.
India is no exception. India is striving hard to achieve energy security. As part of its own energy transition, it has pursued ethanol blending in petrol to serve two stated aims, reducing the crude import bill and raising farmer incomes.[1] However, the way the State has accelerated a fuel transition well ahead of the ecosystem meant to support it has raised serious constitutional implications. This blog sets out what that cost amounts to for a single owner, on figures the government has itself admitted, and then asks whether the decision to place it there meets the standards the Constitution sets for it.
A. What the Transition Costs, and to Whom
(i) A Transition in a Hurry
Ethanol blending is traceable to the National Policy on Biofuels, 2018, which set a considered, long-term roadmap for reducing import dependence and supporting farm incomes.[2] The shift to biofuels was planned well in advance, drawing on the experience of other countries. Even so, a settled policy framework cannot cure a rushed execution. NITI Aayog, in fixing E20 as a 2030 target, proceeded on the premise that the surrounding infrastructure and the vehicle fleet would be ready by then.[3] India, however, brought the change forward. The nationwide rollout of E20 was completed by April 2025, roughly five years ahead of the 2030 target, and E20 became the default fuel throughout the course of that year. The consequence is a mismatch that the government itself now manages. Millions of vehicles built and calibrated for E10 are on the road, and the fuel they were designed for is being withdrawn as the default. The ecosystem was meant to catch up to the fuel. The fuel arrived first.
(ii) The Evidence, and the Evidence Withheld
The VAHAN dashboard records roughly 44.53 crore registered motor vehicles, of which 24.97 crore are active,[4] but that headline count mixes every fuel type and vintage together and does not by itself show how many vehicles were built for the lower blends. Read instead by fuel type and vintage, the same records show around 235 million petrol vehicles registered between 2011 and 2023, the window closing at the April 2023 compliance date, that were made for blends below E20.
An independent estimate points the same way. Thomson Reuters Foundation analysis has put the number at around 234 million vehicles, more than 80 per cent of those sold in the window, compliant only with 5 and 10 per cent blends.[5] Neither count has been tested against actual survival or scrappage, since both measure vehicles registered rather than those still on the road, so each is indicative rather than exact. What no public source yet provides is a count adjusted for how many of these vehicles are still on the road. More pointedly, the ARAI durability testing that would show what each of them actually loses on E20 remains unpublished, and that is where the evidence is genuinely withheld.
On the question of mileage, the government relies on controlled laboratory testing by the Automotive Research Association of India (ARAI), which estimates an efficiency loss of 1 to 6 per cent. Against that admitted range, consumer experience runs higher. A LocalCircles consumer survey drawing over 22,000 responses found that nearly two-thirds, around 66 per cent, of owners of pre-2023 petrol vehicles reported a mileage drop of more than 10 per cent, and that over half reported unusual wear and tear or an increased need for repairs.[6] Both sets of figures must be read with caution. The ARAI testing is conducted under controlled laboratory conditions on a limited set of vehicles, and these conditions do not necessarily reflect the performance of an ageing, heterogeneous on-road fleet. The LocalCircles survey records what owners report rather than what has been measured, and its respondents came forward on their own. However, what is significant is that the two points are in the same direction, and that the government continues to rely on its laboratory range while declining to disclose the underlying study against which the survey findings could be tested.
The government has sought to justify the mandate by citing studies conducted by Indian Oil Corporation Limited (IOCL), ARAI and SIAM, which were cited in a press release of the Ministry of Petroleum and Natural Gas as the scientific basis for the transition.[7] When a citizen sought access to these studies under the Right to Information Act, seeking the underlying research reports of those studies, disclosure was refused on the ground that the reports were confidential.[8] What is known of the ARAI report's findings has emerged only in fragments. The underlying reports and the durability test data, however, remain unpublished. The details on which any independent assessment would depend, including the extent of that component deterioration and the conditions under which it occurred, have therefore never been placed before the public.
A government cannot rely on a body of research to justify a binding measure while withholding that research from the citizens the measure binds. This, more than anything, is why disclosure in the present context is not merely desirable but obligatory.
(iii) The Cost Is Transferred, Not Erased
The government's own figures claim more than ₹1.9 lakh crore in foreign exchange saved and over 310 lakh metric tonnes of crude oil displaced since 2014-15. The policy has also generated more than ₹1.6 lakh crore in additional earnings for farmers. The government further credits it with reducing carbon emissions.[9] The savings are real. What the figures do not show is what the owner of a vehicle built for a lower blend has paid since 2025, and will keep paying for the life of that vehicle. No such figure has been published.
The government has itself conceded a drop in mileage. This admitted mileage drop may be marginal for an individual vehicle owner. That drop, when multiplied across a fleet of many millions, becomes a very large aggregate cost, and this cost has been silently shifted onto each owner of an E20 non-compliant vehicle, in the form of higher fuel consumption and repair bills. This silent cost falls hardest on the commuters and gig workers for whom the vehicle is a means of livelihood.
The approach taken here is to quantify the burden at the level of a single owner, where every input is public and verifiable. The efficiency loss is modelled at two levels, 6% being the upper end of ARAI's official 1 to 6% estimate, and 10% being what owners report, taken from the LocalCircles survey of June 2026 in which 66% of pre-2023 petrol vehicle owners reported a mileage drop of more than 10%. Annual running is fixed at 10,000 km at an assumed baseline efficiency of 12.5 km/l, giving a base consumption of 800 litres a year, with the pump price held constant at ₹100 a litre, against a prevailing rate of ₹102.12 (New Delhi Price). The base year is 2026, the first full year of loss, E20 having become the default during 2025. The 15-year figure is the registration life of a private vehicle counted from the date of purchase. On these assumptions the owner who bought a car in 2022 bears the loss for the longest remaining term, roughly 11 years, giving a cumulative burden of ₹52,800 to ₹88,000. The figures make no adjustment for movement in the pump price or for the declining purchasing power of the rupee over the period.
The same exercise for a two-wheeler, keeping every other variable unchanged, uses a baseline efficiency of 50 km/l, taken as the mean of the common 45 to 55 km/l commuter range, and annual running of 5,000 km, giving a base consumption of 100 litres a year. A 6% loss adds 6 litres a year, ₹600, and a 10% loss adds 10 litres a year, ₹1,000, at ₹100 a litre. Borne over the same longest term of 11 years for a 2022 vehicle, the cumulative burden runs from ₹6,600 to ₹11,000.
Scaled across the surviving legacy fleet, these per-owner figures give a sense of the whole. The count is taken on a deliberately narrow and stated basis. It covers two classes only, two-wheelers and private cars registered as light motor vehicles, which between them make up the overwhelming bulk of the petrol fleet. It counts pure-petrol vehicles alone, leaving out those that also run on CNG or LPG, since the mileage penalty bites only in petrol mode, and it excludes the 2011 cohort whose fifteen-year life ends in 2026, with each earlier cohort carried only for the years it has left. On that basis the cumulative burden on the fleet, weighted by the remaining life of each vintage, runs from about ₹1.23 lakh crore at the ARAI-based 6 per cent to about ₹2.06 lakh crore at the owner-reported 10 per cent, across the years to 2037. A saving credited to the national account is thus set against a burden of comparable size that falls on individual owners and is recorded nowhere.
Further, the cost does not fall evenly across income groups because the legacy fleet consists of 179.1 million, or 91.1 per cent, are two-wheelers and cars make up the other 8.9 per cent. It falls mostly on two-wheelers, which are owned largely by lower-income users. For them, a fixed penalty on fuel efficiency or maintenance takes a bigger share of what they earn.
In addition to the mileage drop, the ARAI research found that several of the rubber and plastic materials from which fuel-system components are made, the NBR and PVC blends, epichlorohydrin and PA66 used in seals, gaskets, O-rings, hoses and pipes, deteriorate more on prolonged exposure to E20 than to conventional petrol.[10] That deterioration, attributable to the change of fuel, brings forward the replacement of these parts, at an estimated ₹500 to ₹2,500 for a two-wheeler and ₹5,000 to ₹10,000 for a car, a cost that may recur over the life of the vehicle.[11] The extent of the repair cost cannot be stated with precision in the absence of the test report. The deterioration itself is not speculative. ARAI has documented it, and the government has not published the data that would measure it. This cost is borne by the vehicle owner.
There is a further issue of proof. Without data distinguishing E20-related damage from ordinary deterioration, long-term harm traceable to the fuel may be brushed aside as normal wear and tear, and warranty or insurance claims declined on that footing. This opacity leaves the owner uncertain of his loss and unable to enforce his rights through the ordinary courts.
Nor is there an easy exit. CNG is no ready substitute. Its refuelling network is far thinner than that for petrol, and its stations often draw long queues. Switching an existing vehicle to CNG involves extra expenditure and may lead to warranty and insurance complications. The push toward electric vehicles brings its own difficulties, including uneven charging infrastructure, urban conditions such as broken roads and monsoon waterlogging that the technology tolerates poorly, and the plain fact that a resident of a multi-storey building often has nowhere to charge in any season. Directing such a person away from petrol does not offer a choice. It relocates the same burden to a different corner of the household.
B. The Question the State Has Not Answered
(i) Right to Livelihood
For a very large share of those affected, the vehicle is not a discretionary comfort but a working asset. Consider the two-wheeler of the platform delivery rider, the car of the app-based driver, and the daily conveyance of the salaried commuter, the small trader, the vendor such as the milkman, and the farmer in the countryside, all of whom an incomplete public-transport network leaves with no real alternative.
(ii) Doctrine of Proportionality
The E20 mandate comfortably clears the first two limbs of the doctrine of proportionality. Energy security and a lower crude import bill are unquestionably legitimate aims, and blending domestic ethanol is rationally linked to them. The rollout becomes genuinely vulnerable, however, when tested against the third limb, necessity, the least intrusive means test. A less restrictive alternative was plainly available. The government could have kept E10 on sale through separate, clearly marked dispensers for the older fleet, while scaling E20 up for the newer, E20-compatible vehicles now being sold. That arrangement would have advanced the same national objective, since ethanol would still be blended and imports would still fall, while sparing owners of older vehicles the financial burden the current design places on them. When the same end could have been reached by a route that inflicts materially less hardship, the harsher route strains the necessity limb.
Nor can the objection be one of feasibility. The government cannot be heard to plead operational or logistical difficulties in making E10 available separately, when fuel stations across the country already maintain separate tanks and dispensers for more than one grade of petrol. If the infrastructure to dispense multiple grades already exists, the burden lies on the State to show that a parallel E10 line was genuinely unworkable, and not merely less convenient or less profitable, before a compatible fuel is withdrawn from those who depend on it.
Proportionality does not ask the judge to run energy policy. It asks whether the government took the least harmful road to its own destination.
(iii) Doctrine of Legitimate Expectation and Manifest Arbitrariness
The doctrine of legitimate expectation is the obvious place to look and it does not carry this argument. In Sivanandan C T v. High Court of Kerala, a Constitution Bench held that legitimate expectation is not a legal right and cannot serve as an independent basis for judicial review, and that an expectation must yield to a larger public interest.[12] The reduction in crude imports is in the public interest, precisely the kind that has defeated such claims before, and owners of pre-2023 vehicles were never promised that E10 would remain available.
The value of the decision lies elsewhere. The Constitution Bench held that in a system rooted in the rule of law, a decision taken without any basis in principle or rule is unpredictable and therefore arbitrary. It concluded that the principles of good administration require the decisions of public authorities to withstand the tests of consistency, transparency, and predictability, to avoid being regarded as arbitrary and violative of Article 14. Quoting an earlier decision, the Constitution Bench accepted that an authority may depart from past practice, but held that any such deviation must be predicated on greater public interest or harm, and that this is the mandate of Article 14.
The distinction that follows is between two decisions treated as one. Blending ethanol to twenty per cent has a basis in the rule. It rests on notified specifications and a stated policy of reducing crude imports, and a court will not review its wisdom. Withdrawing the lower blend from the pump rather than retailing both grades is a different decision, and no basis in principle or rule has been offered for it. No comparison of the two options has been published. No assessment of what the withdrawal costs the owner of a vehicle certified against the earlier specification has been placed in the public domain. The objective is stated. The choice is not.
The manifest-arbitrariness enquiry turns on the abruptness of the withdrawal. Removing E10 without a preserved option for the legacy fleet is not simply unfair to those who relied on the old regime. It is manifestly arbitrary under Article 14 because it imposes a serious burden without any rational, graduated basis.
A challenge on this ground has already reached the Supreme Court. In Akshay Malhotra v. Union of India[13], a petition seeking continued availability of ethanol-free petrol for older vehicles was dismissed at the threshold in September 2025. The dismissal turned on the framing of the claim as one of consumer choice. The stronger question remains unengaged, whether a blanket removal of E10, when a labelled option could have been preserved for the legacy fleet and allowed to lapse as that fleet ages out, satisfies the least-intrusive-means limb. That lapse is not a promise but arithmetic. A private vehicle registration runs fifteen years, so the last pre-2023 vehicle leaves the road by 2037, and demand for E10 falls away with the fleet until no vehicle needs it. The burden of keeping it available is temporary and shrinking by design. It is on that question, and not on consumer preference, that a future challenge is likely to stand or fall.
In this climate of confusion and uncertainty, one disclosure would now close the gap. The count of affected vehicles is already available from the VAHAN records. What remains is the ARAI report, which alone would show what each vehicle loses in efficiency and in component life. That report is held by the same government whose policy imposes the loss, so its withholding is a choice rather than an incapacity. Transparency, not opacity, should be the norm, and a burden imposed without the test the report would allow cannot meet what Article 14 requires.
The reconciliation is not utopian. Jurisdictions with far longer ethanol histories, among them the United States, Brazil and the European Union, make lower-blend or ethanol-free grades available alongside higher blends, with clear labelling at the pump so that consumers can match fuel to vehicle knowingly. A proportionate transition would have done the same here, keeping a labelled option for the legacy fleet and phasing the change, so that the cost of greener mobility did not fall on those for whom the vehicle is not a comfort but a working asset.
Endnotes
Available at: https://www.pib.gov.in/PressReleasePage.aspx?PRID=2159409®=48&lang=2 (last visited on July 3, 2026). ↑
Available at: https://mopng.gov.in/files/uploads/NATIONAL_POLICY_ON_BIOFUELS-2018.pdf (last visited on July 5, 2026). ↑
Available at: https://www.niti.gov.in/sites/default/files/2021-06/EthanolBlendingInIndia_compressed.pdf (last visited on July 6, 2026). ↑
Available at: https://analytics.parivahan.gov.in/analytics/ (last visited on July 20, 2026). ↑
Available at: https://www.context.news/net-zero/in-data-indias-vehicle-boom-poses-challenge-for-biofuels-target (last visited on July 11, 2026). ↑
Available at: https://www.localcircles.com/a/press/page/ethanol-petrol-survey (last visited on July 11, 2026). ↑
Available at: https://www.pib.gov.in/PressReleasePage.aspx?PRID=2283118®=48&lang=2 (last visited on July 12, 2026). ↑
Available at: https://www.cartoq.com/car-news/india-e20-fuel-rollout-rti-research-data-confidential/ (last visited on July 12, 2026). ↑
Available at: https://www.pib.gov.in/PressReleaseDetail.aspx?PRID=2268671®=1&lang=1 (last visited on July 7, 2026). ↑
Available at: https://www.autocarindia.com/industry/e20-could-damage-rubber-parts-in-fuel-systems-of-e10-vehicles-arai-440150 (last visited on July 12, 2026). ↑
Available at: https://www.business-standard.com/industry/auto/using-e20-petrol-in-old-vehicles-maintenance-costs-may-rise-up-to-10-000-126062200342_1.html (last visited on July 12, 2026). ↑
2023 INSC 709, Writ Petition (Civil) No 229 of 2017, decided 12 July 2023, paras 42 to 44. ↑
Writ Petition (Civil) No. 813 of 2025, order dated September 1, 2025 (SC). ↑
Author Vidit Agrawal is a Faculty at IIM-Rohtak and a PhD Scholar at University of Delhi & Deepak Sharma is an Assistant Professor and Head of Department (Law) at Dr. Rajendra Prasad National Law University, Uttar Pradesh. Views are personal.