Earlier this month European Sleeper announced its first route – Brussels-Amsterdam-Berlin-Prague – and the news was gushing. Euronews covered the story, saying how it was driven by flight shame, TimeOut talked of a “a ton of cool new sleeper trains“, and Travel+Leisure talked about a “major upgrade” of night services.

Don’t get me wrong, European Sleeper is a great effort, bordering on the heroic – trying to get a 4 country new night train running is some feat. I wish Elmer van Buuren and the team running it all the very best, and as it connects my home (Berlin) with one of my workplaces (Brussels) I am going to be all too happy to use it.

The problem instead is another one.

The scale-up of night trains in Europe is not nearly ambitious enough.

The only established publicly owned rail firm serious about night trains is Austria’s ÖBB with the investments in its NightJet fleet – increasing its initial order of 13 new trains to 33. Within a couple of years ÖBB will run a bunch more night services, including stepping up services through Germany and Switzerland (more here). Including current stock ÖBB is going to have 50-60 night trains in its fleet.

Beyond that it’s pretty thin gruel regarding new services. SNCF, at the insentience of Macron’s government, is re-instating 2 night train routes – Paris-Nice and Paris-Hendaye. Snälltåget is expanding its seasonal night train from Sweden via Denmark to Germany. RegioJet has added a seasonal night train from Czechia to the Croatian coast, and RDC runs the new Alpen-Sylt night express. Moonlight Express is likewise trying to launch night services from Brussels onwards to Germany or Czechia. All of these operators are scrambling around trying to find second hand rolling stock for these services, much of it over 40 years old and with a maximum speed of just 160km/h.

Above all there is an almost complete absence of Europe’s big rail players – Deutsche Bahn, SNCF, Trenitalia, PKP – in this night train market. The most DB and SNCF are capable of is to let ÖBB run a few services on their tracks, and in SNCF’s case grudgingly budge when the French government demands some night trains.

If we set all of that against the backdrop of what rail firms are doing with their long distance daytime fleets it looks pretty miserable. Deutsche Bahn is in the middle of the delivery of 137 ICE-4 trains (to add to the current fleet of 240+ ICEs) and a further 30 more Velaro units will come later. SNCF has ordered 100 next generation TGVs from Alstom, Trenitalia has ordered 87 Frecciarossa 1000 trains.

That gets us then to the crux of the problem – the companies that have the will to expand their night train routes are lacking the means. ÖBB can stretch to 33 new trains, but the likes of Snälltåget or RegioJet have to make do with second hand stock, and that severely limits their ability to run night train services.

The companies that do have the means – SNCF, Deutsche Bahn or Trenitalia – are not interested in ordering new stock or running night trains as doing so would cannibalise their more lucrative daytime trains business. And the distance a night train covers means crossing a border in Europe is pretty much inevitable, and that means stepping on the toes of some other national champion on the other side of the border.

So what is the solution here?

There are potentially a few.

One solution would be to reduce track access charges for night trains, making them more lucrative to run (although whether this would tempt DB, SNCF or Trenitalia into this business is doubtful). An alternative would be for the EU to procure a fleet of night trains, and lease these to operators, thereby freeing the operators of up-front costs (although that one would be lobbied against hard by SNCF, DB etc). A further route would be to introduce a Public Service Obligation model for night trains in European rail, perhaps as part of a Europatakt, where a competitive tender system could be used to determine which firm would operate a route (although this one would be a major step away from the liberalisation approach pursued in the EU until now). Theoretically the EU could I suppose directly support the operation of services that are not profitable, at least initially, but this too would be a major departure from the politics pursued by the European Commission until now.

As I see it the debate that night trains are a good thing has been won in Europe now. The next question is how to scale their provision. And at the moment the players in the market who could massively scale up night services do not want to, while those who do have the will nevertheless lack the means.


  1. Very to the point article.

    I would just point out that at least domestically, Trenitalia runs quite a large network of night trains. Though they are only using existent stock and I don’t know of any plans to replace its ageing carriages.

    PKP, too, have a lot going on within Poland.

    But as for international night trains, there is little to be hoped unless there is a big injection of new rolling stock from somewhere.

    • Thanks, I know. And in Sweden and Romania there are quite some as well. I didn’t want to write a complete overview of all the services there are at the moment – just about the struggles of starting new ones.

      • Hi Jon, thanks for a very the point article. Scalability, or rather the difficulty in achieving that, is a very serious challenge in raising money, which in turn is making scaling up very difficult. Chicken and egg. Also fully agree to ‘those who want to, can’t, those who can, don’t want to’. How do we get from nice words (European Year of Rail, Green Deal, Digitalisation, lalala) to concrete, positively supporting, actions, specifically from the side of the member states?

  2. If the EU outright buying trains is unpalatable, perhaps it could be structured as a loan arrangement: nominally, the operator buys the train sets with EU-provided money and pays them off over time, and only if they go under and default, the EU takes possession of the trains and leases them out or sells them on.

  3. Peter Forrest

    It would good to see their return, although motorail looks unlikely given the vombo of low cost flights + cheap car rental. Pity, because Fréjus-Calais and Florence-Calais motorail services were a great way to end a European motoring holiday.

  4. Indeed, if a significant share of transportation shall be shifted from air to rail, much more rolling stock suitable for overnight travel must be either manufactured or created by refurbishment of existing seated railcars.

    In order to increase both capacity and efficiency, new rolling stock should be designed for universal use in night trains, daytime trains or extra long night-&-day routes. A proposal, how such a carriage could look like is published here: http://buschbacher.at/railcar.html

  5. Gonzalo Bertrán Sitjar

    In the fourth paragraph, you wrote “insentience”, I believe you meant “insistence”.

  6. Pingback: Saturday round-up – 22nd May 2021 – Are We Having Fun Yet?

  7. Pingback: Cross-border rail in 2023 – Sam Thursfield

Leave a Comment

Your email address will not be published. Required fields are marked *