Railroads versus highways: how a choice from the 1950s makes Brazil spend up to 3x more on freight

In physics and transport engineering, there is a relentless law: the friction of a steel wheel on a steel rail is drastically lower than the friction of a rubber tire on asphalt. This means that when you need to transport large volumes of cargo (grains, ores, fuels, industrial containers, cellulose, or fertilizers) over distances greater than 400 or 500 kilometers, the railroad and waterway beat the truck hands down in energy consumption, cost per useful ton-kilometer (TKU), carbon emissions, and safety.

All the other major continental countries on the planet (United States, Canada, Australia, China, and Russia) designed the backbone of their economies by obeying this law of physics. Brazil was the only territorial power to make the opposite bet: today, about 62% to 65% of all Brazilian cargo travels on tires over highways, while railroads account for only about 18% to 21% (and even this share is concentrated by more than 75% solely in iron ore from two large mining companies).

How did we arrive at this distortion that makes everything from the bag of cement to food in the supermarket and industrial components more expensive?

1. The 1950s choice: short-term won over long-term

In the early 1950s, Brazil had about 37,000 to 38,000 kilometers of railroad tracks. It was a network with severe problems: many lines had been laid out in the 19th century just to connect coffee farms to the nearest port, without national integration and with different gauges (distances between the tracks of 1.00 meter in some regions and 1.60 meters in others, the physical equivalent of trying to run two incompatible operating systems without an integration API).

When the country accelerated its urbanization and industrialization in the 1950s (especially during the government of Juscelino Kubitschek and the subsequent decades of the 1960s and 1970s), rulers had two options:

1. The structural option (high initial Capex, low future Opex): standardize the gauges, rectify winding routes, and build large railway trunks integrating the interior to the coast.
2. The quick inauguration option (low initial Capex, extremely high future Opex): abandon deficit railway branches and open thousands of kilometers of paved highways, which could be inaugurated within a single four or five-year term, while simultaneously attracting the automotive industry and consuming diesel refined within the country.

Brazil chose the second option and went further: between 1960 and 1995, the state-owned Rede Ferroviária Federal (RFFSA) deactivated and ripped up more than 8,000 kilometers of tracks. Today, Brazil has about 30,000 kilometers of railroads (less than it had in 1958!), and only about a third of this network operates with high commercial density. Meanwhile, the United States operates more than 220,000 kilometers of extremely high-efficiency cargo tracks.

2. The bill arrives at the port: up to 3x more expensive

What is the real price of carrying a continental economy on the backs of trucks?

For years, comparative studies by the United States Department of Agriculture (USDA) and the Esalq-Log (USP) group showed a fascinating paradox: inside the farm, the Brazilian rural producer in Mato Grosso or Goiás is more productive and has a lower production cost per ton of soy and corn than the American farmer in Iowa or Illinois.

However, when the soy leaves the farm to travel the 2,000 kilometers to the ship, the tables turn. In the United States, the grain goes down in barge convoys via the Mississippi River waterway or in trains of over one hundred cars to the Gulf of Mexico for a fraction of the price. In Brazil, traveling two thousand kilometers by truck to Santos or Paranaguá has historically cost two to three times more per ton than internal American freight.

This logistic overcost does not affect only agribusiness. In industry and e-commerce, Brazil spends about 12% to 15% of GDP on total logistic costs (transportation, storage, losses, and inventory in transit), compared to about 8% in the United States and the OECD. This difference makes up more than R$ 220 billion annually of our Brazil Cost.

3. The route correction and the technical debt analogy

clientHeight The good news is that in the last decade, economic rationality has started getting back on track through private investment:

• Consolidation of the North-South Railway and the Northern Arc: connecting the heart of Brazil (Goiás, Tocantins, Mato Grosso, and Maranhão) to the ports of the North (Itaqui, Barcarena, Santarém, and Miritituba via the paved BR-163 and waterways) has already drastically reduced the cost and export distance to Europe and Asia.
• Early renewals and new railroads (FICO, FIOL, and Ferrogrão): the use of railway concession renewal grants to build new segments such as the Central-West Integration Railway (FICO) is unblocking historical bottlenecks without depending on the federal public budget (which has 91.4% of primary expenses tied up).

For those who work with software architecture and systems engineering, the history of Brazilian railroads and highways is the most perfect physical illustration of the concept of technical debt:

When you need to deliver a system quickly and choose an improvised architectural shortcut (pulling a dirt road without thinking about the scale of millions of requests), you save time and money in the first month (low Capex). But when the volume grows tenfold, keeping that makeshift fix running with server trucks, manual rework, and daily incidents (explosive Opex) costs three times more than it would have cost to design the right tracks from the start. Both in countries and in technology companies, infrastructure shortcuts always charge compound interest in the future.

Continue deepening: read also Basic sanitation and the delay until 2070: the most basic infrastructure of productivity and X-Ray of the Brazil Cost: the R$ 1.7 trillion annually that drain our competitiveness. Also explore the Agenda Brazil 2050 with an interactive simulator and the Brasília Arcade.

---

References and verifiable readings

• Infra S.A. / EPL (Ministry of Transport). National Logistics Plan (PNL 2035): Cargo Transport Matrix in Brazil.
• CNT - National Transport Confederation (2025). Railway Transport Panel and CNT Highway Research.
• Summerhill, W. R. (2003). Order Against Progress: Government, Foreign Investment, and Railroads in Brazil, 1854-1913. Stanford University Press.