The Metro Train Question: India Built the World's Third-Largest Network — But Is It Working?

Sometime in 2025, India quietly crossed a milestone that would have seemed fantastical a generation ago: more than 1,000 kilometres of operational metro rail, the third-longest urban rail network on earth, behind only China and the United States. The speed of the build is genuinely historic. In 2014, India had just 248 km of metro across five cities; by 2026, it has over 1,070 km across some 21 systems, with daily ridership above 1.1 crore and annual central allocations that have grown five-fold to around ₹30,000 crore. Most of this happened after 2010 — Bengaluru opened in 2011, Mumbai in 2014, Chennai in 2015, Kochi and Hyderabad in 2017, followed by Lucknow, Nagpur, Pune, Ahmedabad, Kanpur and more. In infrastructure terms, the metro boom is one of modern India's great construction achievements.

But construction is not the question anymore. The question of 2026 is harder: are these systems actually solving India's mass-transit problem at scale — and can they survive their own economics?

The utility scoreboard: real, but far below the promise

Start with the honest ridership picture. Studies across Indian corridors find that most metros carry only 25–35% of the ridership their project reports projected — and some far less: Lucknow runs at about 9% of its forecast, Jaipur under 8%, Kochi around 15%, Hyderabad around 15%. Even Delhi, the flagship, had to cut its own projections repeatedly — from a forecast of 1.57 crore daily riders to around 41 lakh in revised estimates — though it now genuinely carries around 60-65 lakh a day and has visibly reshaped how the capital moves. Bengaluru has climbed past 9 lakh daily; Pune touches 2.5 lakh.

So the fair verdict on utility is double-edged. Where networks are dense, connected and mature — Delhi above all — the metro has become indispensable: fast, safe, air-conditioned dignity for millions, especially women, and unquestionably a brake on what pollution and congestion would otherwise have been. But in most cities, the metro remains a partial artery in a body that still moves by two-wheeler, bus and auto. A line or two through a sprawling city, with poor last-mile connections, fares above bus tickets, and stations far from where people actually live, cannot by itself dent traffic on Indian roads. India's metros are succeeding as engineering and underperforming as networks — because a metro only becomes mass transit when it is woven into everything else.

The money scoreboard: losses almost everywhere

Financially, the picture is starker. Metro economics worldwide are unforgiving — hardly any system on earth recovers its capital cost from fares — but Indian metros struggle even at the operating level once debt enters the ledger. Delhi Metro is the honourable exception that proves the difficulty: it earns operating surpluses, yet carries enormous Japanese soft loans it will service for decades. Nearly every other Indian metro runs net losses after interest and depreciation, kept alive by state equity, viability-gap funding and patient debt.

Hyderabad: the cautionary tale written in full

No city tells the story better than Hyderabad — because Hyderabad ran the boldest experiment of all: the world's largest public-private-partnership metro, built and operated by L&T with roughly ₹20,000 crore invested, on the bet that fares plus real estate could make a metro pay. In 2026, the experiment formally ended. After posting a loss of ₹625.88 crore in FY 2024-25 alone, with cumulative losses crossing ₹6,600 crore, L&T transferred 90% of its shareholding to the Telangana government for a payment of ₹1,461 crore, with the state assuming debt of about ₹13,500 crore. From May 1, 2026, the state runs the metro.

Why did the model fail? The arithmetic was merciless: about 4.8 lakh daily riders against an original projection of around 17 lakh — roughly 15% of forecast — producing revenues of ₹1,100 crore a year against annual interest payouts alone of ₹1,273 crore on ₹12,500 crore of borrowings. No operator outruns that equation. Behind the numbers lay familiar causes: last-mile connectivity that never materialised, station real estate and advertising that was never fully monetised, a 169-day Covid shutdown from which ridership never fully recovered as work-from-home took hold, and a city that kept sprawling away from the alignment. There is irony in the geography too: Hyderabad, second in India's metro league by network length in 2014, has slipped to ninth — even as the state now proposes a 163-km Phase 2 across eight corridors, this time as a centre-state joint venture, with the L&T handover itself partly driven by Delhi's insistence that expansion needed a single integrated operator.

Hyderabad's lesson is not that metros are bad. It is that metros cannot be run as private profit ventures in Indian conditions, and that whoever builds them — public or private — must plan for the revenue reality, not the DPR fantasy.

Burden or asset? The honest framework

So are metros becoming a burden on governments? On a narrow accounting view, yes: almost all lose money, and the losses now sit on public books — Hyderabad's ₹13,500 crore being only the newest entry. But the accounting view is the wrong lens, and every serious transit economist says so. Metros are public goods, like roads and schools; nobody asks whether a flyover "makes profit." The proper test is economic return: hours not lost in traffic, petrol not burnt, air not poisoned, accidents not suffered, land values created, and cities kept workable as they double in population. By that test, a well-used metro repays society several times over — and even an under-used one beats the alternative of surrendering cities to cars. The genuine worry is not loss-making metros; it is empty metros — capital locked in corridors that carry a tenth of their capacity while buses starve for funds. The burden is not the metro. The burden is the metro built without the ecosystem that fills it.

The way forward

Which points to the agenda for the next decade — less about pouring concrete, more about filling trains. First, finish the networks: ridership grows non-linearly as lines interconnect, as Delhi proved; half-built systems are the worst of both worlds. Second, fix the last mile as a funded, designed component of every project — feeder buses, shared autos, cycle lanes, walkable station areas — not an afterthought. Third, integrate: one ticket (the National Common Mobility Card), one app, one timetable across metro, bus and suburban rail. Fourth, monetise land honestly — transit-oriented development and station real estate on the Hong Kong model, done transparently, is the only proven route to financial health. Fifth, price the alternative: parking fees and congestion charges that make the metro the rational choice, with the proceeds funding it. Sixth, match the mode to the city — full metros for the megacities, lighter MetroLite and Metro Neo for tier-2 towns, rather than prestige projects destined for 8% occupancy. And everywhere: fund buses too, because the metro's best friend is the bus that brings it passengers.

India's metro story since 2010 is neither triumph nor folly — it is a magnificent skeleton awaiting its circulation. The tracks are laid; the harder, cheaper, less glamorous work of connecting them to Indian life is what will decide whether the world's third-largest network becomes its third-most-used. Hyderabad has shown how the story goes wrong. Delhi has shown how it goes right. The next decade decides which of the two becomes the Indian norm.