Beyond network size

26 June 2026
← back

All too often when discussing complex phenomena, a single statistic rises as the sole measure of fitness or success. Good examples of this are countries' economic development, with GDP as the protagonist metric, or people's cognitive capabilities with IQ as the one and only measure. This simplifying tendency leaves out crucial details and can lead to ill-fated conclusions.

High Speed Rail (HSR) networks, the topic of this article, often suffer from this tendency. In their case the size in km seems to be the only thing that matters.

We add three metrics to the discussion: how much is the network used, how expensive was it to build and how much help came from the EU. We focus on the four major HSR networks in Europe.

Call me a nerd, but I find this image fascinating: very much a loot of research questions waiting to be picked up by curious minds.

To me, three observations seem most intriguing:

1. Spain received more EU funding to build their HSR network than all the other three countries combined.
2. Italy's HSR network costs around 33 mill. € per km, twice as much as any of the other countries.
3. France's network is two thirds the size of Spain's, yet it is used four times more.

I will now dive into each of these three topics by giving additional context and having a go at plausible explanations. Nevertheless, a warning is in place: these are rather complex issues and what follows is far from the last word.

Spain: the spoiled child?

Spain received more EU funding to build their HSR network than the other countries combined. How much exactly? 2.92 times the combined EU funding of France, Germany and Italy. In fact Spain received ~50% of all the money spent in HSR by the EU ¹, which ended up paying for 26% of the network's cost ². But, where does this tremendous asymmetry come from?

Some might point out that Spain has traditionally been economically worse off than its European peers, and therefore funded more generously by the EU.

Nevertheless, in recent years this has not been the case. In 2023 it received less than the other three countries ⁶. And while in the past Spain did generally receive more funding (e.g. 2005: 1.08 times France's and 1.2 times Germany's) ⁷, this figures fall extremely short of explaining a funding 10 times that of France or 5.22 times that of Germany.

I offer another plausible explanation.

Spain's network was developed during the late 90s and early 2000s. This was a time when the EU had gotten very involved in rail transport, with the vision of an integrated European rail network (e.g. 1991's Railway Development Directive).

Spain was a substantial obstacle in this pursuit.

Unlike Italy, Germany or France, its network was extremely outdated ² and it ran on wider than standard rails (i.e. the Iberian Gauge). In other words: it had an old network that was very expensive to upgrade to modern standards and that, even then, could never integrate with the rest of Europe due to its gauge.

In this scenario, the development of a new HSR network of standard width was a good approach towards European rail integration. Brussels probably saw it that way, and thus the funding.

Italy's burden

Building HSR costs twice as much in Italy as it does in the other three countries. It costs 33 million euros per km while in France, the next most expensive, it is less than 15.5 million euros.

Part of this difference actually comes from the other three networks being relatively cheap. Expanding our perspective we see that Japan (~29€/km), Korea (~34€/km), Taiwan (~40€/km), Netherlands (~44€/km) and Austria (~30€/km) all have similarly high costs ⁵.

Now the question flips: why is building HSR so much cheaper in Spain, France and Germany than in Italy? The answer for the first two may be simple: lower population density. This not only lowers land acquisition costs, but infrastructure costs as well. The more crowded the area the harder it is for lines to follow simpler cheaper paths, and more complicated and expensive solutions must be adopted ³.

Germany's case is different, though, since it has an even higher density than Italy. The explanation here lays in their radically different approach to HSR. Their InterCity Express (ICE) trains were integrated into the preexisting rail network. Upgrading sections to allow for the >250km/h speeds. This approach has advantages and disadvantages, but is certainly cheaper.

Another part of this difference does come from Italy having some very expensive lines ⁵. But what led to this?

In the literature one can find explanations referencing poor planning leading to enormous overruns (i.e. 200% initial estimate) ⁴, as well as uncompetitive award of contracts (i.e. +14-20% total costs) ³. The former is, though, the norm rather than the exception. All European lines audited by ECA in 2018 showed an overrun of at least 25% ¹. The latter explanation may have some merit. Nevertheless, it cannot explain a 100% difference with its peers.

Italy also made some very expensive design choices when planning their network ³. To make HSR more "palatable" to smaller towns and regions, they added many extra km of connections (i.e.around +15%) to allow trains to reach local stations. And to comply with other pressure groups the network was designed to also suit heavy freight, leading to the necessity of more tunnels and viaducts (i.e. in certain lines ~5 times more than in Spain or France).

Explanation of the kind "Italy did X which increased the cost" are rather unsatisfactory, though. It is easy to find reports stating "France/Spain/Germany did Y which increased the cost". Making a number of costly decision is not enough of an explanation, since every country does. The compounded effect of these decisions must be estimated and compared to see if they can be considered the cause of this difference. Assessing this, though, is rather difficult and requires substantial field expertise. The question of why Italy has some very expensive lines remains, thus, inconclusive.

France's achievement? Spain's failure?

Before delving into this discussion, a methodological note. We use passenger-km (pkm), a standard measure of passenger use in the field of transportation. It shows the total number of km traveled by passengers during a given period of time, 2023 in our case.

The data comes from three sources. Eurostat for Germany and France, the Comisión Nacional de los Mercados y la Competencia (CNMC) for Spain, and .italo for Italy. We do not use Eurostat for the last two countries because it fails to provide usage statistics including private operators, a substantial market share in their case. Lastly, Italy's data refers to 2019 for two reasons: it was the highest quality data I found and, due to the COVID-19 pandemic, the network use numbers of 2019 are generally very similar to those of 2023.

Ok, let's move on now.

France's network has 2,628km, around 65% of Spain's 3,870km. And yet shows a yearly usage of 64.57 billion pkm, around 380% of Spain's 16.9 billion.

This enormous difference can hardly be explained by France being a larger or more populous country. It is only 9% larger and has 40% more population than Spain.

So what makes the French network so much more effective than the Spanish one?

The inefficacy of Spain's HSR network is an undisputed fact. In 2020 the Independent Authority for Fiscal Responsibility (AIReF) estimated the social return of this infrastructure to be low or negative, even when considering a 50 year time horizon ⁸ . And the literature ^1,2,4 often blames political factors overshadowing technical cost-benefit analyses.

Nevertheless, it is hard to identify the specific design choices that make this network so much less effective at moving the countries' population around than the French one. And I find this topic interesting enough to deserve a whole other analysis and post. So there you go, a bit of a cliffhanger. Stay tuned.

Sources