Europe Day 2026: from 232 to 69 night train connections

Liberalization: where have we come?

For freight trains, market forces have meanwhile become the almost absolute rule; the same applies to international passenger trains. For national passenger trains, market forces are still rather exceptional or, in some countries like Belgium, entirely non-existent. The Belgian train network forms an integrated system, with scheduled trains connecting with one another as best as possible, usually from six in the morning until twenty-three in the evening. Most trains also do not make a profit, and thus the entire SNCB network is actually one OSP, established in the "management contract" that the federal government concludes with the SNCB.

There is an administration responsible for ensuring that passenger needs are met—in other words, that as many connections as possible are provided between municipalities, as quickly as possible, and at the lowest possible price. This administration studies how it can further improve this system by, for example, laying more tracks or switches, or increasing speed in certain areas.

The fact that international passenger trains have been liberalized implies that there is no administration at the European level studying such a plan, but that the free market will meet the needs of travelers. In other words: if enough travelers want to go from point A to point B and are willing to pay for it, the free market should ensure that such a train is created because a railway company sees a profit in it. There is no pressure from any government whatsoever.

Offer of night trains: goal not achieved

Since the year 2000, this system has gradually replaced the old traditional system. The old night trains, which were implicitly cataloged as a public service, disappeared. The Netherlands was the first country to streamline its night trains in 2002, with Belgium following in 2003.

That liberalized system has now been in effect for more than twenty years, and the number of international night trains in Europe fell from 1,257 train pairs per week to 445 per week between 2001 and 2019. The number of different connections decreased from 232 to 69 during the same period. In the meantime, those numbers have declined even further.

The number of traditional daytime trains, however, remained stable during the same period: it went from 2,545 train pairs per week to 2,804, and the number of connections rose from 197 to 200. At first glance, that might not look so disastrous, but compare those figures to the growth in the number of flight connections during the same period, or to the growth in the number of passengers or trains in domestic traffic, and the only conclusion can be that the liberalization of train services has not achieved its goal.

Declining offer despite liberalization

Why this decline in offer ?

We believe it is hardly possible to make a profit on night trains, which is necessary within a liberalized framework. After all, no company will launch a night train without a profit.

It might be possible to make a profit on a few very busy routes—who knows, perhaps it will work on the Paris - Brussels - Berlin route, for example, the train recently relaunched by the Dutch-Belgian operator European Sleeper—but it certainly will not be possible on enough routes to achieve a Europe-wide network, which is our goal.

The night train involves a host of costs that make such a train inherently more expensive than a simple day train: higher personnel costs, because safety must be guaranteed during the night, breakfasts and linens must be distributed, and moreover, the number of passengers per carriage is usually smaller than in a normal seated carriage: after all, people need to sleep and have a lot of luggage with them. There are attempts to increase the number of passengers per carriage—start-ups Nox and Lunarail, for example, are working on this—but so far without concrete results. In any case, even if the attempts are successful, it will still take years before night trains with such a configuration traverse our continent, and even then, only one of the handicaps will have been resolved: high personnel costs remain.

Solution for declining supply: act like national trains with a ‘public service’

If we cannot make night trains profitable, perhaps we should hold a societal discussion on whether we should recognize night trains as a public service and create an OSP for them. We at Back on Track Belgium believe this is desirable.

One twelve-carriage night train can easily transport 400 passengers comfortably in beds over 600 to 1,500 kilometers in one night. That is the equivalent of two Airbus A320 flights saved per train. With about twenty such night trains departing from Brussels every day, forty flights per day are saved, which represents a quarter of the flights from Zaventem. With a range of 1,500 kilometers, one can reach most European cities and tourist regions between dinner and breakfast. The societal benefit seems evident to us.

Carriages: an overview

The fleet of European night trains is very diverse. While some recent carriages are in operation, it is undeniable that the average age of night train carriages is very high.

Carriages fifty years old are not uncommon, and there are outliers of over seventy years old. These include carriages mounted on antique Minden-Deutz bogies, without air conditioning or without emergency brake bypass. This bypass is the system that allows a train driver to temporarily bypass an emergency brake, only to come to a complete stop at a safe location, for example beyond a tunnel. These are not uncommon in night train rolling stock.

Reclining cars, the less comfortable form of sleeping carriages, have hardly been ordered in recent decades. Sleeping carriages, which are more comfortable, have been ordered in limited numbers, particularly by operators with night trains that fall partly or fully under public service. A public service contract (OSP) provides a guarantee for making such a substantial investment. The Bulgarian railways, the Polish railways, or the Austrian ÖBB have made such investments.

Nevertheless, the situation looks less favorable than a few years ago: the large ÖBB order of 33 sets of 7 carriages each has been scaled back to 24 sets of 7 carriages. In France, a large order for new couchette and sleeping carriages was also planned, but there, the order seems to be completely off the table for the time being. Instead, the Corail couchettes, now four decades old, would be renovated yet again.

So, what should such a carriage look like?

If night trains cannot be made profitable and no OSPs are issued for them, then investing in such carriages is particularly unattractive. Whereas leasing companies frequently invest in locomotives or freight wagons, so far no leasing company is willing to venture into investing in night train rolling stock. There is no successful precedent, and therefore leasing companies consider such an investment too risky.

Moreover, it is difficult to manufacture a carriage that meets all requirements. The requirements for new carriages are laid down in TSIs, Technical Specifications for Interoperability.

These TSIs are established by the European Union Railway Agency (ERA) and can be freely consulted by anyone on the internet. The problem is that one can build a carriage that complies with all TSIs, but one is still not allowed to operate it on many lines, because countries often impose additional requirements for, for example, tunnels and bridges. Previously, there was a mutual agreement between railway companies that carriages complying with the "Regolamento Internazionale delle Carrozze," or RIC for short, were automatically allowed entry into any country. Unfortunately, this standardization came to an end in the mid-1990s when various countries began imposing additional requirements. While these requirements do benefit safety, the fact that each country holds different views or places different emphases has led to deregulation resulting in a completely unmanageable tangle.

The ERA is currently unable to change this, and consequently, a carriage capable of running across Europe without problems remains a virtually impossible task and a risky investment.

And what about travel information and tickets?

Regarding travel information and ticket sales, we are waiting to see what the European Commission will publish. In fact, they are expected to present plans soon to somewhat improve the situation again: a "Passenger package".

Here too, deregulation has created an inextricable maze: information regarding the various trains is scattered across countless actors. European Commission President Ursula Von Der Leyen promised to ensure that we could once again book a trip involving different operators on a single ticket. That would bring us back to where we were twenty-five years ago, and that is certainly welcome.

This would primarily imply that passenger rights would apply to all train journeys. Currently, that is not the case: in most instances, you have to book tickets for different trains either on the websites of the respective operators or on booking websites, but these also always sell only a limited range.

Operators are currently not obliged to share their entire range, or even anything at all, with such websites, and if you have tickets from different platforms, the risk of a missed connection lies with you, the traveler, and not with the operator. This The maze is clearly not to the traveler's advantage, and the single-ticket train travel promised by the European Commission is therefore very ambitious. Such a thing would be particularly complex in terms of information technology, but absolutely necessary in our view. A simpler alternative is to extend passenger rights to all train journeys, even if they consist of tickets from different platforms. The Franco-German author Jon Worth advocates for this, and in our opinion, he is right. We are curious to see what the European Commission will produce…