NCMC One Nation One Card vs London Contactless: City Transit Payments

At a Bengaluru metro station in March 2023, RBL Bank and the metro corporation announced that the city's roughly six lakh daily riders would move from a closed-loop metro card to a RuPay National Common Mobility Card. Three years on, the card in a commuter's wallet can technically pay for a bus in Surat, a metro in Pune and a toll on the expressway between them. Whether it actually does depends on gates, readers and operator contracts, and that gap is the real story of transit payments in Indian cities.

Urban transit is the densest, most frequent payment most city residents make. It is also where governments have the most direct control over the payment experience, because they own or regulate the gates. This piece compares India's chosen model, a bank-issued card with an offline wallet, against the paths taken by London, Singapore, Hong Kong and Seoul, and looks at what the now-closed Smart Cities Mission delivered for municipal payments.

The model India chose

The NCMC was launched on 4 March 2019 when the Prime Minister opened the first phase of the Ahmedabad metro, according to NFCW's report at the time. The design has three parts: a RuPay debit, credit or prepaid card with an offline wallet stored on the chip; an indigenous automatic fare collection system called SWEEKAR; and a standard fare gate called SWAGAT. NPCI sets the card and terminal standards, and C-DAC set the fare collection standards, according to an NPCI white paper from September 2023.

The offline wallet is the clever part. A gate does not need to call a bank server to approve a ride. It debits a small balance on the card itself, which keeps tap times low and lets the system work where connectivity is weak. The trade-off is that the user has to top up that wallet, which brings back some of the friction of a closed-loop card. The same card can also be used online like a normal debit card, so the bank account, not the transit operator, is the customer relationship.

By September 2023, 48 issuers had put more than 164 million NCMC-capable cards into circulation, according to the NPCI paper. Acceptance had spread to metros in Kochi, Noida, Nagpur, Delhi (all lines compliant by June 2023), Chennai and Bengaluru (April 2023), Mumbai lines 2A and 7, Kanpur and Pune, along with buses in Ahmedabad, Surat, Goa's Kadamba service and Mumbai's BEST. More than 25 banks were authorised to issue the cards, per ClearTax's guide.

Cards issued is the wrong number

The 164 million figure flatters the system. Many of those cards are standard RuPay debit cards that banks made NCMC-compliant on reissue, and a large share of holders have never loaded the transit wallet. The more meaningful metrics, which are rarely published, would be the share of transit taps made with NCMC on each network and the number of cities where one card works on every mode.

Bus networks are the hard case. The NPCI paper notes that Bengaluru and Chennai buses carried about 35 lakh and 36 lakh passengers a day respectively in 2018-19, far more than their metros. Retrofitting thousands of buses with validators that handle an offline wallet, sync with a back office and survive Indian roads is a procurement problem, not a payments problem. Metros, with fixed gates and central control rooms, were always going to move first.

A transit card is only national when the cheapest bus in the smallest city accepts it, and that is a hardware budget, not a policy announcement.

How other cities did it

The global history of transit payments is a progression from proprietary stored-value cards to bank cards and phones. Each city's starting point shaped where it ended up.

City	First contactless system	Current model	What it shows
Seoul	Upass closed-loop card, 1995	T-money (from 2004), compatible across about 60 cities	A dominant closed-loop card can become a de facto national standard
Hong Kong	Octopus stored-value card, 1997	Octopus remains central; transactions in about 0.3 seconds	Speed and ubiquity can keep a closed loop competitive for decades
Singapore	Magnetic fare cards 1990, EZ-Link contactless 2002	SimplyGo account-based ticketing from 2016	Shifting value from the card to a back-office account
London	Oyster, then bank-card contactless on buses (2012) and Tube (2014)	Open loop dominant for pay-as-you-go	Bank cards can overtake a transit card within a few years
India	Closed-loop city cards, Kochi and Ahmedabad NCMC pilots in 2017	NCMC bank-issued card with offline wallet, from 2019	A hybrid: bank-issued, but still a stored balance on the chip

The Asian dates come from the NPCI white paper's international review. London's numbers are the most instructive. Transport for London introduced bank-card contactless on buses in 2012 and the Tube and rail in 2014. By April 2018, contactless accounted for half of pay-as-you-go journeys on the Tube and rail and 45% on buses and trams, up from about a quarter two years earlier, with about 17 million journeys a week and roughly one in eight made with a phone, according to Digital Transactions. By September 2021 the contactless share of Tube pay-as-you-go trips had reached 70% and about two-thirds on buses, per Mobility Payments.

London's key move was making the bank card behave like a transit card: daily and weekly fare caps applied automatically, with no top-up and no wallet. Users did not have to learn anything. India's NCMC asks users to understand a separate offline balance on a card they already carry, which is a small but real barrier.

Worked example: why tap speed matters

Consider a busy metro gate. Hong Kong's Octopus processes a tap in about 0.3 seconds. At that speed, the card transaction itself could in theory support about 200 taps a minute; the real limit is how fast people walk through the gate. If a card instead needed an online authorisation taking, say, 1.5 seconds, the same gate would cap at about 40 taps a minute from the payment side alone.

Now apply that to a Bengaluru-scale network with about six lakh daily riders, most of whom pass through a gate twice. That is roughly 12 lakh gate transactions a day, concentrated in two peak windows. A one-second difference per tap, multiplied across a peak hour, decides whether a station needs more gates or longer queues. This is why both NCMC's offline wallet and London's deferred-authorisation model exist: neither waits for a bank's real-time approval at the gate. The difference is that London settles the fare against the bank account afterwards, while NCMC deducts from a pre-loaded balance.

What the Smart Cities Mission delivered for payments

The Smart Cities Mission, launched in June 2015 for 100 cities, formally ended its funding phase on 31 March 2025. By 11 July 2025, 7,636 of 8,063 projects, worth Rs 1,53,977 crore, were complete, and the Centre said the mission was financially closed with no budget for the current year, according to a Rajya Sabha reply reported by ETV Bharat. Cities had claimed Rs 47,459 crore of the Rs 48,000 crore central allocation.

Every one of the 100 cities now has an Integrated Command and Control Centre, and the mission added over 84,000 CCTV cameras, more than 1,740 km of smart roads and intelligent transport management systems, according to a Down To Earth review, which also found that only 18 of the 100 cities had completed every project by March 2025. What the published progress data barely mentions is payments. Municipal fee collection, parking and property tax were folded into city apps in many places, but there is no national data on how much municipal revenue moved to digital channels as a result. The command centres watch traffic; they mostly do not run the cash.

That is a missed opportunity. Transit, parking and municipal fees are the payments a city government controls most directly, and a common card or a common UPI collection layer for all of them would have been cheap compared with road works. Our explainers on UPI and prepaid payment instruments cover the rails a city could use, and India's digital public infrastructure shows why the national layers are already in place.

Open loop, UPI or a digital rupee?

India now has three plausible ways to pay a fare. NCMC is built for the gate. UPI dominates small payments everywhere else, but a QR scan at a turnstile is slower than a tap, and UPI's offline variants are still small. The e-rupee has been piloted with offline features, which we discuss in offline CBDC payments. Our view is that the gate will stay card-based for years because tap speed beats everything, while the top-up and the account behind the card are where UPI and possibly the e-rupee will compete. For a broader comparison of faster payment systems, see instant payments around the world.

What to watch next

• July to August 2026: Parliament's monsoon session. Written replies are the likeliest place for updated NCMC issuance and acceptance data by city, which has not been published comprehensively since 2023.
• Through 2026-27: the remaining Smart Cities projects in states such as Uttar Pradesh and Haryana, several of which were due by December 2025 or April 2026, and whether cities fund command centres once central money ends.
• 1 February 2027: the Union Budget for 2027-28, which will show whether the Centre funds a successor urban programme and whether transit payment hardware for city buses gets dedicated money.
• 31 March 2027: the end of the financial year by which several metro projects under construction are expected to open new lines, each a chance to launch with full NCMC and open-loop acceptance from day one rather than retrofitting.

NCMC issuance figures cited here date from September 2023, the latest comprehensive public count we could verify. Check NPCI and metro operators for current acceptance.

About this researchMuffinPay InsightsIndependent analysis of stablecoins, CBDCs, cross-border payments and digital public infrastructure. Every figure is linked to its source. Read our editorial standards →