Xanh SM vs Grab: The Electric Taxi War in Vietnam’s Cities

The electric taxi war in Vietnam is the only place in the world where a car maker’s founder built a ride-hailing company to buy his own cars, and then took the market leader’s crown with it. Green and Smart Mobility, the company behind the Xanh SM brand, was created by Phạm Nhật Vượng with his own money, launched in Hanoi in April 2023 with a fleet of cyan VinFast cars, and within twenty months was reported to have edged past Grab in car bookings. That result is genuine and so are the qualifications: Xanh SM competes with the advantage of cheap capital, cheap cars and no need to make a profit on the timeline that a Singapore-listed platform must observe. This article explains how each side’s model works, what the market-share numbers actually measure and what the fight means for the wider mobility sector. It is part of the Vietnam Company Stories hub.

How did Grab come to dominate Vietnamese ride-hailing in the first place?

By arriving early, buying its main rival and adapting to the motorbike. Grab entered Vietnam in 2014, made GrabBike its core product because the country rides two wheels, and in 2018 absorbed Uber’s Southeast Asian business, leaving it as the only scaled platform in the market.

The Uber deal was decisive. Uber had been in Vietnam since 2014 and had been losing money on subsidies to drivers and riders in a fight it could not win regionally, and when it sold its Southeast Asian operations to Grab in March 2018 for a stake in the combined company, Grab inherited Vietnam’s second-largest network overnight. Vietnam’s competition authority investigated the transaction, concluded that Grab’s post-merger share exceeded the notification threshold, and ultimately fined it only modestly; the practical result was a near-monopoly in app-based rides in the two big cities. Gojek, the Indonesian rival, entered in 2018 under the GoViet brand, never got beyond a distant second and withdrew from Vietnam in September 2024. Be Group, a locally founded platform backed by VPBank and later by other investors, has hung on as a credible third player, particularly in two-wheel rides.

Grab’s model is capital-light in the classic platform sense. It owns no vehicles, classifies its hundreds of thousands of Vietnamese drivers as independent partners, takes a commission of roughly 20 to 30 percent on each fare, and layers food delivery, parcel delivery, and a payment wallet, GrabPay by Moca, on the same customer base. It has also been a relentless lobbyist: the long fight over whether ride-hailing should be regulated as a taxi service or a technology platform, resolved in 2020 by Decree 10, which required app operators to meet taxi-style obligations, was fought largely between Grab and the traditional taxi companies Vinasun and Mai Linh, who sued Grab for unfair competition and won a small judgment. By 2022 Grab’s Vietnamese business was profitable at the contribution level, according to the group’s regional disclosures, and the market looked settled.

Why did Phạm Nhật Vượng start a taxi company?

Because VinFast needed customers and Vietnam’s streets needed proof that its electric cars worked. GSM was founded in March 2023 with initial capital of about VND 3 trillion, roughly $125 million, and a stated plan to buy 10,000 cars and 100,000 scooters from VinFast and put them on the road as taxis and rental vehicles.

The timing tells the story. VinFast had listed on Nasdaq via a SPAC merger in August 2023, was reporting enormous losses and had just abandoned petrol cars for an all-electric range that Vietnamese buyers were slow to adopt. A charging network was still being built, residual values were unknown and the brand had a reputation for early quality problems. A large, visible, professionally run electric taxi fleet solved several problems at once: it gave VinFast guaranteed volume in a year when consumer demand was thin, it put tens of thousands of people into VinFast cars as passengers who would not otherwise have tried one, and it generated real-world durability data that the company could point to. Vượng funded GSM personally rather than through Vingroup, which kept the losses off the listed conglomerate’s accounts and avoided a further related-party burden on Vinhomes shareholders.

The related-party nature of the arrangement is disclosed and is large. VinFast’s filings show that GSM and other Vượng-affiliated entities accounted for a majority of its deliveries in 2023 and a substantial share in 2024, and that GSM has bought or leased on the order of tens of thousands of VF e34, VF 5 and VF 8 cars plus a fleet of Feliz scooters. Analysts who cover VinFast treat Xanh SM demand as a distinct category from retail demand, because it is; the taxi company’s purchases are a decision by the founder about how to deploy his capital, not evidence that a stranger in Hanoi chose a VinFast over a Toyota. That said, the fleet exists, carries paying passengers and earns fares, and by 2025 it had become the largest electric taxi operation outside China.

How does the Xanh SM business model actually work?

It is a hybrid of a classic taxi company, a fleet-leasing business and a ride-hailing app, run in three layers: a directly owned fleet with salaried drivers, a franchise and leasing arm that supplies VinFast cars to independent taxi firms, and since 2024 a platform that lets individual VinFast owners take bookings.

The owned fleet came first and remains the core. Xanh SM hired drivers as employees on a base salary plus a share of fares, put them through a training programme on service and on the specifics of driving and charging an EV, and dressed the cars in a single cyan livery. Passengers book through the Xanh SM app, hail on the street, or call a switchboard, and early customer surveys consistently rated the service above Grab on cleanliness, driver behaviour and the absence of surge pricing. The company priced slightly above Grab at launch, then moved to parity and periodic discounts as the fleet grew. Because it owns the cars, its unit economics are those of a taxi company: revenue per car per day against depreciation, energy, driver cost and maintenance, and the electric drivetrain helps on energy and maintenance while the purchase price, even at a related-party discount, is the largest cost line.

The second layer is where the strategy becomes a distribution channel for VinFast. GSM offers electric cars to established taxi companies in the provinces, such as Lado Taxi in Lâm Đồng and Én Vàng in Hải Phòng, on lease or sale terms, sometimes with Xanh SM booking integration, and it has signed such deals with dozens of operators. The third layer, the Xanh SM Platform launched in 2024, invites private VinFast owners to register as drivers, extending the network without further capital. Abroad, the company launched in Laos in late 2023, Indonesia in December 2024 and the Philippines in 2025, always with VinFast cars and usually with local partners, giving VinFast an export channel that does not depend on dealers. Every layer sells cars; the taxi fares are the return on the cars once sold.

What do the market-share numbers really measure?

Only the car segment, and only bookings, not revenue or profit. The widely cited estimate that Xanh SM held about 37 percent of four-wheel ride-hailing in the fourth quarter of 2024 against Grab’s roughly 36 percent came from a market-research firm, is contested by Grab, and says nothing about the motorbike and delivery segments where most trips happen.

The composition of the market matters more than the headline. Vietnamese ride-hailing is, by trip count, mostly two-wheel: GrabBike, Be Bike and their rivals move far more passengers than cars do, at fares of a dollar or two, and food and parcel delivery on motorbikes generate a further large slice of platform revenue. Xanh SM entered the scooter segment with Xanh SM Bike in late 2023 but remains a small player there, and it has no food-delivery business. Grab’s Vietnamese gross merchandise value across all services is on the order of a billion dollars a year, and the portion exposed to Xanh SM’s strongest position is the higher-value car segment, where it hurts, but not where the volume is.

Even within cars, the comparison is between unlike things. A Xanh SM booking includes street hails and phone bookings for a company that is, in regulatory terms, a taxi firm; a Grab booking is a platform transaction on which Grab keeps a commission and the driver keeps the rest. A share point for Xanh SM represents fares flowing through an owned fleet with a very different cost base. Independent analysts have tried to estimate Xanh SM’s economics from the cars it has bought, driver headcount and typical utilisation, and most conclude that it is loss-making at the operating level once depreciation is charged at market rather than related-party prices, which is consistent with the company’s own silence on profitability. The honest summary is that Xanh SM has won a large share of the segment it chose, at a cost that only its founder knows.

How has Grab responded?

By defending price in cars, accelerating its own electrification and leaning on the parts of its business Xanh SM cannot reach. Grab cut fares and commissions selectively, introduced an electric-car option using VinFast and other vehicles driven by its own partners, and expanded delivery, advertising and financial services where it faces no cyan cars.

Grab’s first move was the obvious one for a marketplace: it lowered the price of GrabCar in the cities where Xanh SM was strongest, reduced commissions for drivers at risk of defecting and ran loyalty promotions through its wallet. Its second was to remove the environmental argument. It signed agreements with vehicle financiers and with electric-vehicle suppliers, including a 2024 arrangement under which drivers could rent or buy VinFast cars for use on the Grab platform, and set targets for the share of electric trips in Vietnam. The irony of Grab drivers in VinFast cars competing with Xanh SM drivers in VinFast cars was not lost on anyone, and it illustrates the deeper point: VinFast sells a car either way, and the founder’s bet pays off whichever platform wins the passenger.

The third response is to grow around the problem. Grab Vietnam’s delivery business has expanded steadily as the country’s e-commerce and food-ordering habits deepened, its advertising business sells placement to restaurants and brands, and GrabPay, in partnership with Moca, competes in the e-wallet war. The company has also continued to invest in two-wheel rides, where its network of drivers, the largest in the country, and the motorbike’s structural role in the motorbike economy give it an advantage that an owned fleet of cars cannot erode. Grab has lost the narrative in cars, but not the business as a whole, and its regional profitability since 2023 gives it the ability to absorb a long fight in one segment of one country.

What happened when Xanh SM went abroad?

It became VinFast’s export channel. Launches in Laos in November 2023, Indonesia in December 2024 and the Philippines in 2025 each put hundreds to thousands of VinFast cars on foreign roads under a controlled brand, before dealers in those countries had sold many, and gave the car maker a story for its investors.

Laos was the trial: a small market next door with weak incumbents, where Xanh SM could test running a fleet outside Vietnam’s regulatory and charging environment. Indonesia was the serious move. It is Southeast Asia’s largest car market, Grab and Gojek’s home ground in two-wheel ride-hailing, and the site of a VinFast assembly plant under construction; a taxi fleet there both consumes the plant’s output and builds brand awareness in a country where VinFast had no presence. The Philippines followed the same logic. In each case the local operation relies on VinFast for cars, on partners for charging and sometimes for licences, and on GSM for capital.

The foreign ventures are also where the model’s limits show. Xanh SM’s advantage in Vietnam is the founder’s willingness to fund losses and the related-party price of the cars; abroad, the cars still arrive at that price but the operating costs, the regulatory friction and the incumbents’ strength are all higher, and there is no domestic groundswell of support for a national champion. Fleet sizes overseas remain small relative to Vietnam, and the ventures are better understood as marketing and volume for VinFast than as ride-hailing businesses with independent prospects. Whether that changes depends on whether VinFast’s Indonesian and Indian plants produce cars cheap enough for a taxi operator to earn a return on them without a subsidy.

What does the electric taxi war mean for the wider mobility sector?

It has pulled Vietnam’s electrification timeline forward by years, forced every platform and taxi company to plan for electric fleets, and demonstrated that charging can be built at scale by a private group without waiting for the state. It has also shown that vertical integration can beat a marketplace in a segment, at least while capital is patient.

Before Xanh SM, electric cars in Vietnam were a curiosity; by 2025 the two largest cities had tens of thousands of them in daily commercial use, a network of V-Green chargers in car parks, petrol stations and apartment blocks, and an emerging secondary market for used VinFast taxis. Traditional taxi firms such as Mai Linh and Vinasun, which had spent years losing to Grab, found themselves choosing between electrifying with VinFast, electrifying with Chinese imports such as BYD, or continuing with hybrids from Toyota; several chose the last, some the first. The grid implications, including the load on EVN, the state electricity monopoly in cities where distribution networks are already strained, have started to feature in planning documents.

The Hanoi petrol-motorbike restrictions from July 2026 add a second front. Whoever supplies the electric scooters and the charging or swapping for the city’s delivery and ride-hailing riders will hold a large market, and Grab, Be, Xanh SM and the scooter makers are all positioning for it. For the platforms the lesson is that owning assets, which the entire ride-hailing industry spent a decade avoiding, can be a source of advantage when the asset itself is the product a customer wants to try. For everyone else it is that in Vietnam a determined conglomerate with a founder’s backing can create a market segment, take the lead in it and change the regulatory conversation within two years.

What can founders and investors learn from Xanh SM versus Grab?

That a captive demand channel can be a legitimate strategy when the product needs proof, that asset-heavy models are not obsolete in emerging markets, and that market share bought with related-party capital must be discounted until it survives without it. Both companies are case studies, but in different subjects.

For founders, the Xanh SM case shows how to solve a cold-start problem in a category customers distrust. VinFast could not persuade retail buyers to take a risk on an unknown electric car; a fleet that put a million people a month in the passenger seat, with a trained driver explaining the charging, did the persuading at scale. The general form of the idea, building the first large customer for your own product, is old, and Vượng applied it with unusual commitment. The cost is that the product’s true market demand is obscured, and outside observers, including the equity market, will assume the worst until independent demand is visible.

For investors, the Grab side is the more useful. A marketplace with a strong position in the high-volume, low-ticket segments of a market can lose the prestige segment to a subsidised entrant and still be the better business, provided it does not panic into matching the entrant’s losses. Grab’s decision to defend cars selectively while growing delivery and financial services looks, on the evidence so far, like the right allocation of a finite budget. For operators in any market where a conglomerate arrives with a captive supply chain, the question to ask is what happens to the entrant’s cost base when the parent no longer needs the volume; that moment, whenever it comes, is when the real competition between the two models will begin.

Frequently Asked Questions