Electricity’s share of India’s road transport energy mix rose from 0.3 petajoules in 2016-17 to 33 petajoules in 2025-26, but still accounts for less than 1 per cent of the total.
India’s EV transition began largely with e-rickshaws, which scaled through informal financing before formal policy support shifted attention to two-wheelers, buses and other segments.
E-rickshaws accounted for 48 per cent of all electricity consumed by road transport in 2025-26, showing that the transition is still concentrated in a few vehicle categories.
Delhi stands out for reducing diesel use through court orders, economic disincentives and targeted policy, while most other states have seen electricity replace petrol and CNG more than diesel.
Ever wondered just how important the energy transition is for the transport sector? We have heard the new battery electric vehicle registration numbers and the rising electrification percentages, and it is indeed a success story. But a 35 per cent electrification rate in 2025-26 does not mean 35 per cent of the transport energy mix was electric. At most, and this is a stretch, it means that of the new energy demand added that year, something in that range might have come from electricity.
To understand this properly, we need to look at how the transport energy transition has actually played out across states over the years.
Nationally, electricity’s share of transport energy rose from 0.3 petajoules in 2016-17 to 33 petajoules in 2025-26, a roughly 110-fold increase, faster than petrol, diesel or CNG on a ratio basis. And yet, after a decade of subsidy, policy attention and headline-writing, electricity is still a little under 1 per cent of total transport energy.
Both numbers are true, and they are not in tension. They point to a transition that is real but narrow, spreading fast in the vehicle segments and states where conditions were right, and going nowhere at all where they were not.
Check the dashboard below to see how different states’ transport energy mixes actually transitioned, and their corresponding battery electric vehicle composition.
This raises a natural question. Why did the transition move the way it did, and what does its shape reveal about what is actually driving it?
India’s total EV registrations from 2016-17 read as a clean climb: 55,856 electric vehicles that year, rising to 2.55 million by 2025-26. Look inside it, and it is two different stories stitched together at a seam around 2021.
From 2016-17 through 2020-21, e-rickshaws made up between 62 and 95 per cent of every electric vehicle registered in India. In 2017-18, the peak year, they accounted for 94.6 per cent. This was not a straightforward policy win. FAME-I and FAME-II barely acknowledged the category. E-rickshaws scaled on informal financing and cheap imported lead-acid battery packs, sitting outside most state EV policies entirely, while filling routes no municipal bus ever served.
Then the mix inverted. The share of e-rickshaws in national battery electric vehicle registrations fell from 61.7 per cent in 2020-21 to 36.5 per cent, 30.5 per cent, 30.5 per cent, 26.3 per cent and finally 22.2 per cent by 2025-26. This was not because e-rickshaws slowed, but because everything else finally started moving, and moved faster.
The June 2021 revision to FAME-II, which roughly tripled the two-wheeler subsidy, landed at the same time as petrol crossed Rs 100 a litre and the first serious state EV policies, in Delhi, Gujarat and Maharashtra, came into force. Two-wheeler EV penetration went from 0.3 per cent of new registrations in 2020-21 to 6.6 per cent by 2025-26.
That handover, from an unsubsidised informal segment to a subsidised formal one, is the actual shape of India’s e-mobility transition. It did not begin with policy. It began with the informal market. Policy showed up nine years in.
Registration numbers tell one story. Electricity consumption tells another. Nationally, e-rickshaws, a segment that barely features in the EV policy conversation, account for 48 per cent of all electricity consumed by transport in 2025-26. Two-wheelers, three-wheeler passenger autos and buses each take roughly 12 per cent. Cars take 10 per cent. Cabs, despite carrying much of the narrative weight in discussions on ride-hailing electrification, account for just 3 per cent.
This is a ceiling that charging infrastructure alone cannot lift. Maruti, the dominant cab original equipment manufacturer, does not sell an electric-eligible cab model at all, and the few electric cab options from other manufacturers remain limited and expensive. Three-wheeler goods carriers take under 3 per cent.
If the question is genuinely where India’s transport energy transition is happening, the honest answer is that it is mostly in e-rickshaws, and mostly not in the segments getting the policy attention.
State breakdowns sharpen this further. In Uttar Pradesh, e-rickshaws account for 77 per cent of the state’s transport electricity. Cabs are 0.2 per cent, a figure so small it barely survives rounding. In Bihar and Assam, e-rickshaws take 79 per cent and 71 per cent respectively, with three-wheeler passenger autos as a distant second. Cabs in both states are close enough to zero to round away.
Karnataka is the only state without one segment dominating. Two-wheelers, buses, cars and cabs are all in the 14-29 per cent range, the closest thing here to a transport-electricity mix that is not being carried by one vehicle type or one procurement programme.
Delhi and Maharashtra look different, for a reason worth naming: buses. Buses take 31 per cent of Delhi’s transport electricity and 35 per cent of Maharashtra’s, both far above the national 12 per cent share, because electric-bus procurement under FAME-II and PM-eBus Sewa is concentrated heavily in these two states. It is a tender-driven mechanism, visible here as a genuine share of the energy itself. It is also worth noting that in 2025-26, electricity constituted around 0.3 per cent of total transport energy in Maharashtra, whereas in Delhi it was 2.4 per cent.
But scale is not the only thing worth measuring, and this is where Delhi’s win actually matters. E-rickshaws dominate the electricity numbers almost everywhere, but what they displace is mostly nothing, or at most CNG and petrol on the margin, not diesel.
The point of the transition, if this is a climate and public-health argument and not just a battery electric vehicle scoreboard, is to move fuel out of diesel’s hide specifically, because diesel is where the bulk of transport’s pollution burden sits. Buses and trucks are the one large-volume segment that was genuinely diesel-first. Delhi’s 31 per cent bus share is not just a big number.
Everywhere else, the electricity being consumed is mostly displacing CNG and petrol. In Delhi’s buses, it is displacing both CNG and diesel.
Delhi is the one place where diesel’s retreat looks like policy rather than incidental substitution. It is worth naming why, because the mechanism is different from every other state.
Diesel vehicle registrations in Delhi fell 72 per cent between 2012-13, the peak year for diesel registrations in the capital, and 2025-26. The fall was led by a 96 per cent collapse in diesel cabs, driven by a sequence of legal mandates stretching back to a 1998 Supreme Court order converting all public transport to CNG.
This was later tightened through a rapid Bharat Stage leapfrog to BS-VI and reinforced by Delhi-specific economic disincentives, including an Environmental Compensation Charge on heavy trucks, an Environmental Pollution Charge on large diesel cars and SUVs, an Air Ambience Charge on local diesel sales, and a hard ban on 10-year-old diesel and 15-year-old petrol vehicles.
Delhi’s own transport-energy data confirms the direction. The capital shows the single largest absolute diesel decline of any Indian state, 29.62 petajoules between 2016-17 and 2025-26, well ahead of Haryana in second place.
That is the difference between what happened in Delhi and what happened everywhere else. Elsewhere, the shift between fuels has largely been a market response to charging time, refuelling access and who happens to own a driveway. In Delhi, it was compelled by court order and cost, and aimed specifically at the categories that most needed compelling.
Delhi’s current EV policy carries that logic forward, prioritising buses, two- and three-wheelers, delivery fleets and ride-hailing aggregators over private car incentives.
That raises a further question. Did a compelled transition like Delhi’s spread to its neighbours the way an organic one might?
Between 2016-17 and 2025-26, CNG’s gain rose steadily along India’s gas pipeline corridor, from 5.3 percentage points in Gujarat to 15 in Delhi. The corridor runs through six states: Gujarat, Rajasthan, Madhya Pradesh, Uttar Pradesh, Haryana and Delhi.
Electricity followed no such pattern. Uttar Pradesh, at 1.11 percentage points, sits ahead of Haryana, at 0.29, despite bordering Delhi. No state has caught electrification from a neighbour the way it caught CNG.
The likely explanation is regulatory, not technological. India’s gas regulator licenses areas under a minimum work programme, a station schedule backed by a performance bond. Electricity’s mandate is comparatively weaker. Fuel outlets must carry one alternative fuel since 2019, gas by default and EV charging as a fallback. FAME-II funded chargers as a subsidy, not a penalised buildout. Nothing compels a charger onto a route the way a CNG licence compels a station onto a corridor.
This is worth testing, not asserting. CNG, petrol and electricity may compete with each other more than with diesel, although that is unproven here and inconsistent across segments. Two-wheeler electrification tracks against petrol. Three-wheelers plausibly track against both.
Assam and Meghalaya are the cleanest cases. CNG never gained three-wheeler share, yet EV reached 91.8 per cent and 50.7 per cent by 2025-26 anyway. Tripura and Bihar look different. CNG built real share, 67 per cent in Tripura by 2021-22 and 84 per cent in Bihar by 2022-23, before losing most of it. CNG has since fallen back to 17 per cent and 33 per cent respectively by 2025-26, while EV has climbed to 82 per cent and 60 per cent over the same period.
It is the same pattern, arriving later and finishing further along. The divide looks less like geography, and more like whether CNG ever got a foothold there at all.
Uttar Pradesh is not just an illustration of this pattern but its single largest driver. On a base of 13 per cent of the national three-wheeler market, Uttar Pradesh’s own three-wheeler CNG share peaked at 82.3 per cent in 2022-23 and fell to 23.9 per cent by 2025-26. Its EV share rose from 6.1 per cent to 74.8 per cent over the same period, accounting for nearly 30 per cent of the entire national increase in electric three-wheelers by itself.
Diesel sits outside this fight. Nationally, its volume keeps rising. Trucks, buses and the standing car fleet dwarf what these segments displace. But within the segments where CNG, petrol and electricity actually contest, diesel has been pushed out fast: from 45 per cent to 12 per cent of three-wheeler registrations, and from 55 per cent to 13 per cent of cab registrations, between 2019-20 and 2025-26.
Diesel loses decisively where it is actually contested. It just is not contested in enough of the market to move the national number.
The first answer is charging infrastructure that is affordable, accessible and implementable.
Implementable means extending CNG’s licensed-buildout-with-bond model to EV charging: a real timeline and an encashable performance bond, not a paper target. Right now, a 70-kilowatt commercial connection can take DISCOMs four to six months, and sometimes more than a year, to energise.
Accessible means awarding buildout obligations to dedicated charge point operators, not fleet operators building just enough charging to run their own vehicles. The two are not the same business. A fleet operator’s incentive stops at its own fleet. A charge point operator’s entire commercial model depends on utilisation and uptime for anyone who pulls in. Awarding tenders to fleet operators quietly recreates the ‘one outlet, box ticked’ problem in a new form.
This means targeting hubs at cab ranks, bus depots and resident welfare association parking areas through single-window approval, rather than treating one generic outlet as EV’s fallback wherever gas is not available.
Affordable means tariff clarity and commercially viable charging economics before any mandate. Delhi’s own passenger three-wheeler permit-replacement rule is the clearest evidence of what happens without it. A driver forgoing an old three-wheeler permit for a new one is choosing CNG over EV, because the charging guarantee is not there to make the electric option the obvious pick.
The second answer is making the zero-emission vehicle mandate implementable, not aspirational.
This requires phased, escalating sales targets with real compliance flexibility, including Bureau of Energy Efficiency-administered credits, the way CNG’s own bond was enforceable. It cannot be a number on a policy page that original equipment manufacturers can quietly fail to meet. A mandate only works if it is built to bind the manufacturers currently sitting it out.
It must also be bundled with financing. Battery-rental and leasing schemes can strip out the capital expenditure gap, so compliance does not simply get pushed onto drivers and small fleets as risk.
Delhi’s 1998 diesel mandate remains the clearest proof that a binding, enforced obligation moves a fuel mix faster than a price signal. CNG had both a captive market and an implementable buildout obligation. Electricity currently has a weaker version of one and none of the other.
India did not fail to make electric vehicles affordable. It has not yet made electric vehicles available on the terms its highest-utilisation fleets actually need.