It’s been more than a year since we did a deep dive on Brazil. Since then, the country has closed a strong sales year even though electric two-wheeler sales are not much to talk about.
But then InsightEV likes to identify the green shoots, and Brazil is promising.
Introduction to the Shocking Number
“Let’s say around 40%, or higher.”
I paused. I did not believe him.
The next day I would ping Billy on WhatsApp and recheck. Did Jack say 40% or 14%?
40%.
Jack Sarvary and Billy Blaustein are cofounders of Vammo, the battery-swapping commercial mobility enabler in São Paulo. I am in the middle of a Google Meet interview, and Jack is discussing the actual finance rates for commercial motorcycle riders.
I don’t believe Billy either. Startup founders can be inventive with numbers to make the business case look good. I go on the internet.
The numbers thrown up are pretty much what Jack mentioned. If you are a new motorcycle buyer for non-commercial usage and with a good credit profile, you would be lucky to get finance rates close to 20%. At the other end of the spectrum, a commercial motorcycle rider will end up paying anywhere between 35% and 50% on financing.
That’s a lot. That’s worse than credit card rates in India.
That’s Africa territory.
And that’s what makes Brazil complicated. It is not an easy, comprehensible country. In some ways it is like the Global North. On other parameters, it is not very far from the central belt of Africa.
It is huge
This is what we wrote in May 2025.
The equator passes through Brazil just about 300 km north of Manaus, the capital of the state of Amazonas. The city is significant, and we will revisit it later in this analysis. Because Brazil practically lies on the equator, it is the least impacted by the Mercator distortion, the phenomenon that makes countries further away from the equator appear much bigger on a cylindrical map.
To sum up, Brazil is huge. At 8.5 million sq km, it is the fifth-largest country in the world.
However, don’t let that landmass fool you. More than half of Brazil is uninhabited. It would never be. It’s like Australia.
Not too many people.
Again, from our May 2025 analysis.
However, it’s not very populated, with only about 200 million inhabitants. A great river flows through the country, and nearly two-thirds of the land mass is the rainforest irrigated by the river.
Hollywood tells us that man-eating snakes and venomous insects exist in that forest. Then, the climate army says that the almost uninhabitable rainforest is an area of great ecological importance. If anything happens to it, like too many humans pilfering through it, the glaciers will start melting.
When we said great river, we meant it – the Amazon and its tributaries are the largest freshwater hydrological system in the world. The river is massive, and crossing it is a challenge compounded by flooding – it bulges many times its size in a flood. Understandably, most transportation is up/down the river rather than across it.
As a result, most prominent Brazilian cities are near the coast. Most population density is around the coast, except for some scattered cities, like Manaus, close to the river.
São Paulo is the largest city by far, followed by Rio de Janeiro, the erstwhile capital. The present capital, Brasilia, ranks third.
They are higher in Per Capita GDP (PPP)
At USD 24,428, Brazil’s per capita GDP, PPP basis, is significantly higher than Indonesia at USS 18,973, and is almost double that of India, sitting at USD 12,801. Its per capita GDP, PPP basis, is three times that of Kenya and six times that of Rwanda.
But the wide diversity in Africa means that Brazil is still only around Egypt, and compared to anothr southeast asian country, Malaysia, at USD 46,986, it is still way low.
InsightEV – Global Electric Mobility
They have higher speeds and taller people
“In speaking with people and observing the market, road conditions, driving speeds, and overall traffic behavior, the conditions are much faster-paced than what we’ve seen in India and Southeast Asia,” Billy informs me.
While the country has its own struggles with the quality of roads, the network in the south and east, where most of Brazil’s population is concentrated, comprises some smooth roads, wide highways, and concessions (privately managed toll roads). Average speeds are much higher than in most of the Global South.
Then there is the average male height – Brazil stands at 5’7.5″ compared to India at 5’5″, Indonesia at 5’4.5″, and Vietnam at 5’6″.
Taken together, the above factors explain Brazil’s motorcycle buying decisions.
That income level is why Brazilians ride 160cc motorcycles rather than 110cc ones. We have made this argument before, and it still holds that in large two-wheeler markets, engine displacement tracks per capita income. Roughly half the Brazilian market sits in the 160cc class, and about 30% is 110cc-125cc, the inverse of the Indian distribution. It is also why the electrification maths is harder here than it looks. A wealthier rider is less desperate to save on fuel, and a bigger bike is more expensive to replicate electrically.
The TWAR and FAR are not the most conducive
The Two-wheeler Affordability Ratio (TWAR) is a measure of the prices of the most popular two-wheeler in the market relative to the GDP per capita, PPP basis, of the country.
The Fuel Affordability Ratio (FAR) measures the ratio of pump prices for gasoline to the GDP per capita, PPP basis, of the country.
Countries with high TWAR and FAR are the most conducive to electrification for obvious reasons.
Brazil’s love for 160 cc motorcycles, which come at higher price points than 110 cc/125 cc, means that the TWAR spikes up. Arguably, this also makes it an incorrect measure to use here. The most popular motorcycle in Brazil is the Honda CG150, which retails at about R$20,000, and the TWAR picks up that price. But being the most popular would not make it the default for a delivery rider. They have much cheaper 110cc- 125 cc motorcycles available. TVS and its local importer, Mottu, have created an entire business around that.
At the same time, fuel prices are comparable to most of the Global South, but the higher per capita GDP on a PPP basis reduces the FAR. While there is BoM cost space for a new electric player, there is low attraction for electric motorcycles in the retail market.
An economy that grows slowly and lends expensively
One of the reasons why Brazil’s two-wheeler lending rates are astronomically high is because the benchmark interest rate, known as the Selic rate, sits at 14.0%.
I don’t buy that.
The Indian repo rate sits at 5.25%, and two-wheeler finance starts at less than 8.0%. Even for delivery fleets, business finance rates start at 14.0%.
There is no reason that a Selic at 14.0% should result in two-wheeler finance rates at 35%+. Either the financial institutions are too greedy, or they are scarred by defaults/frauds.
Then there is the problem with the real policy rates. Both countries sit at about 5% inflation, which means the real policy rate for India is zero. For Brazil, it is 9.0%, one of the highest in the developing world. There is a massive room for rate cuts, and the Selic has been coming down. In August 2026, Brazil’s central bank cut the Selic to 14.00%, the fourth reduction in a sequence that began earlier in the year from a peak of 15.00%.
The problem is the slow economic growth. The IMF’s July 2026 update put Brazilian GDP growth at 2.4% for 2026 and 2.2% for 2027. There is a general election in Q4, and the economy may benefit from a pre-election fiscal stimulus.
An economy expanding at 2% a year with a real policy rate of 9% is the single most important fact about the Brazilian electric two-wheeler market. It is more important than the price of petrol, which at USD 1.3/litre is high but not abnormally so. It is more important than the fact that the grid is roughly 90% renewable. It is more important than anything in the vehicle itself.
The actual financing rate of 40%+ available to delivery riders, especially the ones with poor credit scores, is what creates an underlying challenge for the growth of the segment.
It also creates an opportunity for the leasing/renting business.
Everything happens in a handful of cities
The population of roughly 220 million is concentrated on the eastern seaboard, with the notable exception of Manaus, sitting on the Amazon about 300 kilometres south of the equator, to which we will return.
Greater São Paulo alone holds around 23.4 million people across a metropolitan region of nearly 8,000 square kilometres. Jack Sarvary of Vammo makes the point better than any statistic does. They had earlier mentioned that their immediate business plan does not warrant an expansion beyond São Paulo. Asked whether being confined to São Paulo was a constraint, he pushed back on the premise:
“We’re ‘only’ in São Paulo, but São Paulo is actually made up of around 26 different cities that together form the greater metropolitan area. If you compare it to San Francisco, it stretches from San Jose all the way through Oakland, Berkeley, and Marin. In most parts of the world, that wouldn’t be considered one city; it would be considered ten.”
BTW, if you did not notice, that population of 23.4 million is eerily close to the count in Taiwan. We know there are about 3,000 swapping stations across Taiwan. So the potential to grow within São Paulo is immense. From Taiwan to Bengaluru, to Kigali, and Nairobi, we know swapping works when we have small pockets of high population density.
That concentration cuts both ways for anyone building light electric mobility. It makes a swapping network economically viable, because a single metropolitan build-out reaches a rider population larger than most countries.
São Paulo is a monster of a city and anyone mastering that has already learnt how to handle one of the highest crime rates in Brazil, a challenging topography, and high road speeds. It’s a great primer for expanding anywhere else in Brazil. Things should be easy after that.
The delivery ecosystem is being fought over by three continents
Brazil’s app economy is large and formalising rapidly. IBGE’s experimental module on platform work found 1.7 million people working through digital service platforms in the third quarter of 2024, equivalent to 1.9% of private-sector employment and up 25.4% from 1.3 million in 2022. Of those, 964,000 were principally in passenger transport, 485,000 were delivery workers, and 294,000 were in general or professional services.
A parallel Cebrap study commissioned by Amobitec, drawing on data supplied by iFood, 99, Uber and Zé Delivery, put the 2024 total at 2.2 million, with driver numbers up 35% and courier numbers up 18% against 2022.
These figures undercount the addressable market, and the reason matters. IBGE counts principal occupation. Vammo’s own framing is broader:
“In Brazil alone, you’re talking about three to five million people who use their motorcycles for these commercial purposes.”
The gap between 485,000 and three to five million is the difference between people who describe themselves as delivery workers and people who use a motorcycle to earn money. It includes part-timers, independent couriers working outside the platforms, moto-taxi riders, safety patrols, and the informal economy that Brazilian labour statistics have always struggled to see. When Vammo says there are roughly 500,000 app-registered riders in São Paulo alone, and that adding moto-taxi and non-platform couriers takes the figure well above 800,000, that is a market definition, and opportunity, worth taking seriously.
The competitive picture on the demand side changed completely in the last eighteen months. iFood, founded in 2011 and owned by Movile, has historically held more than 80% of Brazilian digital food ordering, a level of concentration with few parallels anywhere. That position is now under coordinated assault. Didi relaunched 99Food in April 2025 with a 24-month zero-commission offer for restaurants, going live in Goiânia in June and São Paulo in August. Meituan signed a five-year, one-billion-dollar commercial agreement during President Lula’s May visit to China and launched Keeta in Santos and São Vicente in October, with a stated commitment of R$5.6 billion over five years. Rappi has pledged R$1.4 billion over three years. iFood responded with a partnership with Uber, a complaint to CADE against 99Food in March 2026, and a São Paulo lawsuit against Keeta and Meituan in May 2026 alleging commercial espionage.
Life in Brazil seems exciting.
For anyone selling vehicles into this market, the litigation is noise and the subsidy war is a signal. Three well-capitalised platforms competing for rider supply means rider earnings are being propped up, rider counts are growing, and the number of motorcycles required to service the demand is rising. Billy Blaustein’s reads it as:
“The market is huge. The motorcycle fleet is growing. App-based delivery penetration is surging. There are more multi-billion-dollar Chinese companies entering the market to do app-based delivery.”
What a rider actually earns
Vammo’s Jack Sarvary estimates that a full-time rider working 60 hours a week across five or six days grosses about USD 1,000 a month, which Jack describes as more than twice the minimum wage.
The favorable comparison to minimum wage understates the rider’s position. If riders today are grossing R$5,000, they are not poor by Brazilian standards, but are running a small business, albeit one with high variable costs. There is definitely a TCO advantage for electric to target.
The credit picture is where they remain genuinely excluded. As Billy Blaustein puts it:
“These are the guys who are not getting approved for vehicle financing. They either don’t have a credit record, or they’re seen as high risk by traditional lenders.”
In short, these guys end up paying the aforementioned 50% finance rates.
Fuel, electricity, and where the savings actually come from
Brazilian pump prices in August 2026 stood at R$6.53 per litre, equivalent to USD 1.28. When InsightEV last examined Brazil in detail in May 2025, the figure was USD 1.12 per litre. Petrol has become roughly 14% more expensive in dollar terms since. It may partly be because of the dollar losing its mind.
Vammo claims that running cost per kilometre is about 75% lower on electricity than on petrol, and that the total package delivers roughly 30% savings against the next-best combustion option. Take a rider covering 150 kilometres a day, six days a week. That is roughly 3,900 kilometres a month. A 160cc commuter returning 35 kilometres per litre burns about 111 litres, or R$725 at current pump prices. Cut the energy cost by 75%, and the rider saves something on the order of R$545 per month. This is close to a third of the federal minimum wage, but not transformational on its own.
Vammo’s 30% figure includes the complete cost of operating a Vammo subscription against the complete cost of running an ICE motorcycle. It takes into account fuel, maintenance, tyres, brake pads, oil, insurance, IPVA (annual vehicle tax in Brazil), licensing, and financing. Vammo’s own framing is that going electric removes IPVA, licensing, and insurance from the rider’s cost base entirely.
Sarvary is emphatic that the riders do this arithmetic themselves:
“Many people have the misconception that delivery riders don’t understand the math very well; that’s not true. They know the economics of their business extremely well. They know exactly how much they spend on gasoline, tires, brake pads, oil, and so on.”
The 40% problem
Ask why Brazil has not electrified, and the standard answers are higher power and top-speed requirements, range, infrastructure, and price. They are all secondary. The primary answer is the cost of money. Vehicle financing for commercial usage sits at around 40% or higher, and that is the rate a captive lender such as Honda would offer to a customer who qualifies. Most delivery riders do not qualify. They buy used motorcycles through informal channels at rates above even that.
That’s a bummer. An electric motorcycle costs more upfront and saves money over time. That trade is attractive when capital is cheap, not so great when capital costs 40% a year. The TCO maths in India makes sense as every rupee of upfront premium on an Indian electric scooter is repaid through fuel savings within a manageable window. In comparison, every real of upfront premium on a Brazilian electric motorcycle has to clear a 40% hurdle rate before the rider is better off.
This is why the Brazilian market went to subscription rather than sale. Vammo considers battery swapping the enabler, not the business.
“Battery swapping is a high-capex, low-margin business, whereas the real economics in vehicles are often on the financing side. So we asked ourselves: why would we bring a bunch of companies to market and let them capture the more profitable part of the value chain while we carried the burden of operating the high-capex, low-margin battery-swapping business?”
That is where Brazil differs from Taiwan or India, and is much closer to Africa.
Similarity to Africa brings the same flavors
“Brazil does not carry the same load as Africa,” Jack says. Add to that, the speeds are higher, the roads significantly better, and one can argue that the machines would need to be higher powered than what you need for Africa.
The high cost of capital remains a common flavour. Like in Africa, this creates an advantage for anyone who is well capitalized and can efficiently deploy capital. For now, Vammo at USD 70 million equity raise and USD 40 million in debt, is leading the race.
Where Vammo actually is
Public reporting through late 2025 had Vammo at 5,000 motorcycles and 150 swap stations, following a USD 45 million Series B led by EIF, the largest electric two-wheeler round in Latin American history. However, Jack mentions that the 400-strong company now runs approximately 10,000 vehicles, and about 1,000 battery cabinets across more than 300 physical locations. Each cabinet has nine slots for eight batteries. That is around 8,000 batteries in the network, plus roughly two per vehicle in the fleet, giving a total on the order of 28,000 packs in circulation. On a Vmoto CPX-class pack, that is somewhere near 90 MWh of deployed batteries.
Vammo provides further growth guidance of 50% by the end of 2026, which would put the fleet around 15,000 and make the company roughly four times larger than at the Series B close, all achieved in about eighteen months.
Beyond that, Sarvary talks about growing ten times from the end-2026 base over four years.
“You’re talking about approximately 150,000 motorcycles over four years in São Paulo alone, focusing only on the motoboy segment. There are roughly 500,000 of those riders in São Paulo alone. So even at those numbers, we’d only be talking about something like a 30% market share if we never left São Paulo.”
That makes it interesting. Even if Vammo just focused on converting São Paulo only, it’s a huge company in the making. Everything else added together would make the case even more compelling.
Mind you, there is ample space for others. Vammo’s USD 45 million was a big round but pales in comparison to what Sun Mobility or Spiro have invested in, or are planning to invest in Africa, or what Gogoro has put down in Taiwan. Even Yulu’s USD 95 million round a few weeks back in India was 2X more than Vammo’s, and that raise was for vehicles, not batteries.
This is more a comment about Brazil than Vammo, and how nascent battery swapping and vehicle renting opportunity still is.
Manaus, and the localisation trap
Almost every motorcycle sold in Brazil is assembled in Manaus. The Free Trade Zone dates to 1957 and was designed to industrialise Amazonas without clearing rainforest, offering exemption from import tax, exemption from federal excise duty, negotiable reductions in state VAT, exemption from federal social contributions, and corporate income tax discounts. Honda, Yamaha, Suzuki, Kawasaki, Bajaj, Dafra, Shineray, Haojue and BMW Motorrad all build there. Abraciclo calls it the largest two-wheeler production hub outside Asia and expects it to produce 2,070,000 motorcycles in 2026, up 4.5% on the 1,980,538 built in 2025.
The benefits are conditional on localisation, and that condition is the barrier no importer can buy their way past. Import duties on complete motorcycles run as high as 104%. Blaustein is direct about it:
“To receive the tax benefits, you need to localize part of the product. That’s what we do today. As we continue to scale volume, we’ll need to localize more and more components.”
Vammo assembles through a partner, Grupo DBS, which is building the Vmoto CPx and CPx Comfort models, with a plan for 15,000 units and around 200 direct jobs.
The supply chain underneath does not exist yet, and Vammo is having to build it.
“In some cases, we’re even training suppliers on how to manufacture components they haven’t built before. In other cases, we’re helping them adapt existing products so they work for our requirements… It’s been challenging, but it’s also a significant barrier to entry that would make it difficult for anyone to copy us,” Jack continues
We flagged last year that Brazil had no local electric two-wheeler component base and expected movement within 12-18 months. What is emerging is not a supply chain that arrived on its own. It is both a challenge and an immense opportunity.
Electric is a very small part of Brazil
Brazilian motorcycle registrations hit 2,197,851 units in 2025, according to Abraciclo, up 17.1% and the best sales ever in the country, beating previous highs of 2008 and 2011.
On the electric side, Fenabrave recorded 8,552 electrified motorcycle registrations in 2025, up 9.4% from 7,821 in 2024. However, through the first four months of 2026, electric motorcycle sales reached 7,502 units, up 33.9% year on year. Statistically, this is a huge growth, all driven by Vammo’s additions to its fleet. However, at a country level, it remains small.
Billy Blaustein’s structural observation about penetration deserves more attention:
“When you compare Latin America with South Asia, and especially Southeast Asia, you see significantly lower motorcycle penetration in LATAM, even when comparing countries with similar GDP per capita.”
Brazil has roughly 18 motorcycles per hundred people. Thailand has something closer to 30 and Vietnam well over 60, at income levels that are not dramatically different.
Billy skips India, and rightly so. India’s per capita GDP is about half that of Brazil, and while India may be the largest two-wheeler market in the world, the penetration at just above 14 two-wheelers per hundred people remains low.
Where do we go from here?
The Selic has to keep falling. The lending rates bother us. At this high rate of high cost of finance, there is no way that electric two-wheelers will take off in retail sales. And it is just not about retail sales. If rates come down, even leasing/lending businesses like Vammo get cheaper capital, which can be passed through.
We love the delivery platforms’ subsidy wars. That will only push motoboy incomes up, increase their numbers, and also electrification. Meituan, Didi and Rappi are collectively committing billions of Reals to take share from a competitor holding 80%. This may be a slightly long-drawn-out fight and presents the best window, right now, in which the delivery fleets grow fastest.
The retail commuter use case is unproven. Vammo has begun retail sales, and Blaustein is explicit that nobody in Brazil is doing for commuters what Ather or Vida do in India. He estimates 15 to 20 million Brazilians commute by motorcycle, 150-200 km a week. The economics for that rider are far weaker than for the motoboy. It’s not convincing, yet Vammo is trying. There is a chance as the finance rates are better towards that part of the customer spectrum.
Then there is the government intervention, or the lack of it. The flex fuel policy means that the Brazilian government has no incentives to offer. There is no subsidy to provide a timely push to electrification, and that makes Brazil genuinely different from India, China, Rwanda, and Vietnam.
That means unit economics would be the primary driver.
Whatever gets built here will have been built on unit economics rather than on public money. That is a slower path. It is also a considerably more durable one.
