Europe's rail freight decline looks different when distance enters the calculation

Rail freight performance in the European Union fell 1.8% in 2025, Eurostat reported on September 14. Its detailed account puts the total at roughly 368 billion tonne-kilometres. The unit matters as much as the percentage: this is a measure combining the weight of goods with the distance they travel, not simply a count of tonnes loaded onto trains.

A worked example shows why the distinction affects the story. Moving ten tonnes for 100 kilometres produces 1,000 tonne-kilometres. Moving the same ten tonnes for 50 kilometres produces 500. The weight is unchanged, but measured transport performance halves. These are illustrative figures, not an explanation of the EU's actual decline; establishing its causes would require evidence about the underlying movements.

Different measures produce different rankings

Germany accounted for 33.5% of EU rail freight performance in the release, followed by Poland at 14.7% and France at 9.3%. The same release gives a separate ranking of tonnes transported per resident, led by Luxembourg, Austria and Lithuania. There is no contradiction. The first comparison measures a country's contribution to total distance-weighted freight, while the second relates a weight measure to population.

A per-capita freight statistic is not a measure of how much rail freight each resident personally uses. Equally, the country with the largest total does not automatically provide the most efficient service. Population, distance and the measure being ranked all affect the comparison. None of these tables, taken alone, establishes delivery reliability, prices paid by shippers or the quality of connections at a particular terminal.

The decline was not uniform

The detailed report places 2025 after annual declines in 2022, 2023 and 2024. It also records declines in each quarter of 2025 against the matching quarter a year earlier, with the second quarter's 5.0% reduction the largest. This gives the annual result some time structure. It still does not identify a single cause, and the data should not be turned into a claim about one policy or disruption without additional reporting.

Check the boundary of the dataset

Reporting coverage is another limit. The detailed system focuses on main railway undertakings, with goods thresholds of at least 200 million tonne-kilometres or 500,000 tonnes, although some countries also include smaller undertakings. Data on consignment type are voluntary. That makes a missing entry in a specialised table different from evidence that no such freight moved. Before making a country comparison, the detailed reporting notes deserve the same attention as the largest number in the table.

Eurostat's transport metadata describes the territoriality principle: transport performance is assigned to the territory where it takes place. For rail, movements on national territory are counted regardless of the nationality of the vehicle. That makes these geographic figures different from a ranking of railway companies by headquarters or ownership. A cross-border operation and a national company can appear in statistics organised around entirely different boundaries.

The detailed freight account separately covers national, international and transit movements, as well as types of goods and consignments. Those distinctions offer a more useful starting point for a follow-up than the aggregate alone. A question about domestic industrial demand calls for different evidence from a question about transit traffic; neither can be answered merely by pointing to the same annual percentage decline.

For readers following supply chains, the practical habit is to keep the unit attached to the number. Tonnes describe weight, tonne-kilometres add distance, and per-capita figures introduce population. The new figures document a fall in EU rail freight performance. Understanding where it happened, what moved and why it changed takes the next layer of data—not a more dramatic version of the headline.