Encouraging train travel over aviation

One idea I've been rolling around for a while now is a way of making non-air travel easier within Europe. For long-haul travel, there isn't really a viable alternative to aircraft - as David JC MacKay points out in Sustainable Energy: Without the Hot Air, ocean liners actually pollute more than aircraft do per passenger mile. But for short-haul travel, especially within Europe, trains are a viable alternative from both an emissions perspective and a time perspective - Europe has an extensive network of high-speed rail.

However, once you dive into the pricing, two issues become apparent: firstly, because of scheduling and other reasons, it's generally slower to go via train (with a number of clunky changes) than by air; and secondly, it's way more expensive. I had some idea of the issues, but wanted to dive into more detail on why the prices are the way they are. I probably won't be able to cover all of it, as it varies by country, but will try and illustrate the main points, and see if there's a potential solution (that doesn't involve massive legislative changes, which are both unlikely to happen and slow when they do).

Is it a bird or a plane?

First, let's take a look at air travel. The high speed rail (HSR) that exists in Europe operates at a range of speeds from around 200km/h up to 320km/h (under 200km/h is generally not considered high-speed rail), whereas a typical commercial jet airliner will cruise at 740-930km/h. Even taking into account slowing down for take-off and landing (more of a factor for short-haul), the aircraft is likely travelling at more than twice the speed of the train. However, that's offset by the extra time costs incurred by flying - security checks will often require being at the airport 1-2 hours prior to departure, and due to their size and noise, airports tend to be well outside of major cities, so there may be another 30mins - 1 hour of travel time involved. DW have a nice graphic of the time required for a flight versus trains (I highly recommend the whole post - it gives a thorough analysis of the differences in the two modes). In short, for relatively short distances (e.g. London-Amsterdam), with all the extra faff, planes actually take longer, but for longer trips (e.g. Paris-Barcelona), planes come out ahead.

Planes also have an advantage in routing and infrastructure. It's immensely expensive to set up an airport, but the journey between two airports that already exist doesn't require much beyond some routing and air-traffic rules. For example, if cities A, B and C all have airports, and you can fly from A to B, and B to C, the cost of setting up a route to fly from A to C is not particularly high. There are network effects - a new airport can connect to all the existing airports, enhancing them and immediately becoming valuable.

But what about costs? Most of us have seen or heard about those £10 plane tickets to insert name of European city break destination here. Trains are definitely not that cheap (more on that in a bit), but how are flights so cheap? The answer lies in a combination of business practices and legislation.

On the business side, most of the cost savings have come from low-cost airlines, who've taken cost saving and efficiency to an art form, perfecting fast turnarounds, and becoming notorious for charging for everything from cabin bags to being able to get on the plane first. They run with very low profit margins, and make up for minimal revenue per flight by making as many as possible. The legislative side is where things become trickier. Airlines don't pay fuel tax, which might come as a surprise to some. Fuel for commercial aircraft in the EU is exempt from taxation, and the 1944 Chicago Convention prevents countries from taxing airlines for the fuel already in an aircraft when it lands. However, EU countries have been allowed to tax fuel for domestic aviation since 2003, but no EU state does. An internal EU report on aviation fuel taxes that leaked earlier this year suggests that even applying the minimum allowed fuel tax of €330 per 1000 litres would increase ticket prices by around 10%, and decrease the total number of travellers by 11%, thus cutting emissions by the same percentage.

International airline tickets are also exempt from VAT, under a loophole that dates back to the formation of the EU, when temporary derogations were put in to exempt international air and sea journeys from VAT, with the rates to be reviewed on regular basis. However, this review never happened, and as new states joined the EU, they demanded the same conditions, so the temporary became permanent. Another part of the same 1977 directive means that the purchase and lease of aircraft and spares, including maintenance, supplies, in-flight meals and more is all VAT-free. This has been estimated to reduce prices for consumers by 15-20% from what they otherwise would be. The story is a little more complex than that, as the aviation industry (in the UK at least) receives few subsidies, whereas rail and bus travel are significantly subsidised. In addition, all forms of passenger transport in the UK are zero-rated for VAT.

As a final blow, whilst the EU Emissions Trading Scheme expanded to cover the aviation industry in 2012, the aviation sector is able to effectively buy its way out of reducing its emissions by purchasing some of the excess permits allocated elsewhere. Aviation also has a static cap, set at -5% of 2004-2006 levels, unlike many other sectors, where caps on emissions decrease year on year. The aviation industry also receives around 85% of its allowance of emissions permits free. This was initially done to prevent carbon leakage (where industries move away from the ETS zone to avoid the costs), but given that the scope of the ETS was reduced from all aviation to intra-European flights, there's little danger of that happening. Aviation emissions in the EU grew by around 5% last year, whilst all other industries fell by nearly 4%.

I like Trains

OK, so we've got a feel for planes, how about trains then?

Trains have an inherent speed limit. As mentioned above, most HSR in Europe operates between 200-320km/h. Experimental trains have been run up to 570km/h, but advances beyond that are likely to require a novel technology, such as Maglevs, which would require entirely new track infrastructure. In addition, high speed rail generally requires dedicated tracks to run most efficiently and safely, so has a high infrastructure cost. This means the cost of linking up two cities with existing stations to a high speed rail network is significant, as new tracks will be required despite the station infrastructure already existing at either end. However, most HSR trains can use conventional rails, meaning that high speed services can run from city-centre stations, decreasing the extra travel time required.

However, every country has their own high speed rail system, with different voltages, signalling, and other differences that make inter-operation tricky. It does happen, but can require trains that can cope with all of this different equipment. The Eurostar Class 373 trains support four operating voltages, two different power collection methods, are able to operate from multiple different platform heights, and support seven different signalling modes. The EU is aiming to develop a Trans-European high-speed rail network, which means that all new high speed rail development at least adheres to common standards, improving interoperability as time goes on.

Ticketing is also complex if you want to travel internationally. Whilst in recent years there have been several international routes opening up (for example Eurostar services from London to various Western European cities), generally each country operates its own system, with separate ticketing. This means longer distance journeys will involve multiple trains and multiple tickets, all purchased from different operators. Many countries at least have some overarching national ticket provider(s) (e.g. TheTrainLine in the UK), but there are few that offer the same service across multiple European countries (lots of operational complexity, technical integration problems, and low margins make it a tricky nut to crack).

HSR has also been accused of pushing out existing slower services, effectively making train travel more expensive and actually slower in some cases, as the new high speed routes don't cover as many places as the previous slower trains.

There are also legislative complexities on the train side too, though this time they tend not to result in cheaper prices for the consumer. Each European country works differently, so I'm going to focus on the UK, as that's where I'm based, and it's the system I'm most familiar with. Train ticket prices in the UK are renowned for being very expensive, and for increasing way above inflation. Why is that? Well, that requires a brief history lesson (as always ;))

The UK Train system

The UK train system, like a lot of UK infrastructure, was privatised in the early 1990s, after decades of public ownership - the idea being that the private sector could run the railways more efficiently, giving better service for a lower cost. The split resulted in three major sectors of the railways being run by different private operators - infrastructure, the rolling stock, and the train services themselves.

The infrastructure was controlled from 1994 to 2002 by Railtrack - a group of companies owning track, signals and other rail-related infrastructure, and most of the UK's stations. After a series of fatal crashes in the late 90s, the regulators became more and more vigilant, resulting in a conflict between Railtrack, who maintained that they were doing their job well, and the regulators, who found repeated evidence that the company was abusing its monopoly position and not maintaining safety standards. After the Hatfield rail crash in 2000, Railtrack completely unravelled, with spiraling costs from both compensation for the crash, subsequent repairs to the network and ongoing maintenance and upgrades plunging the company into administration. In 2002 most of the remaining assets of the company were bought by Network Rail, a government-owned not-for-profit company formed to run the railways, which it has done ever since.

Rolling stock (the trains, carriages and other wagons that run on the rails) are run by Rolling stock operating companies (ROSCOs), which then lease the trains to passenger and freight train operators. Unlike many other aspects of UK railways, the ROSCOs aren't subject to close regulation - they compete relatively freely with one another to provide services to their various customers. However, since the late 1990s there have been complaints that the ROSCOs have been acting as an oligopoly to maintain artificially high prices. After a number of investigations, regulations were changed to try and reduce the power that ROSCOs had over their pricing, and to make sure they compete more effectively.

Passenger rail services in the UK since the privatisation are franchised, with a competitive tendering process to award the various services to train operating companies (TOCs), usually in geographic areas, or sometimes by service type. Franchises generally last for at least seven years, and in theory at least, are non-exclusive, meaning that companies can compete with each other in certain ways. The main costs that the TOCs face are track access and station leasing (both from Network Rail), rolling stock leases (from the ROSCOs) and staffing. They then make revenue from ticket sales.

Franchises are subsidised using a 'cap and collar' scheme, whereby when bidding for the franchise, the TOC must agree a target revenue number. If they achieve more than that, they must pay most of the excess back to the government on a sliding scale (the more they make, the more of the extra they pay back). Similarly, if they fail to meet their revenue numbers, after 4 years they will be bailed out by the government, again on a sliding scale. This approach has been criticised for encouraging the operators to put in reckless bids to win franchises, secure in the knowledge that they'll be bailed out if they fail to make their numbers. The subsidies aren't even across the board - some franchisees are paid large amounts in subsidies, whilst others pay large amounts back to the government.

All this is a long way of saying that the way the railways are run is often complex, and figuring out where the money goes is hard. Fullfact, a fact-checking charity, did analysis to find out if, as often thought in the UK, the TOCs make massive profits. As it turns out, the answer is more complex than that - some companies make profits, and others don't, and it all varies year on year. British trains are reknowned for being expensive, and analysis shows that for standard return tickets, UK tickets are indeed significantly more expensive than elsewhere in Europe, although the UK does tend to have much cheaper off-peak/flexible offerings. In other words, it's cheap if you're flexible about when you travel - not an option for many commuters.

What can we do?

The biggest and hardest nut to crack here is legislative. Creating a seamless inter-European high speed rail network will take time, but if there's passenger demand, it's easier for governments to allocate funding for new infrastructure. However, with airlines receiving massive tax breaks and not paying for the environmental externalities they cause, trains cannot compete on a level playing field. Changing legislation, especially when fighting against an entrenched and powerful industry such as aviation, will take a lot of time and money, and can't be relied upon. So what else can be done?

I definitely think easier access to cross-Europe train tickets would help bring more people to train travel. Yes, the prices currently can't compete, but currently even those who actively choose not to fly have a hard time travelling any other way - there are dedicated travel agents who specialise in dealing with all of the complexities of booking train tickets across different countries. This is an area technology and software could excell in - wrapping all of those providers in APIs, and offering a single, easy-to-use service that gets people across Europe easily. It could even help innovate with pricing - once the service can be shown to bring in revenue, companies would potentially be more open to experimenting with pricing structures. I've also considered a service that prices travel inclusive of carbon emissions - so flights become correspondingly much more expensive. However, ultimately this doesn't help if consumers can simply purchase flights elsewhere for the current (low) prices.

Ultimate high speed rail can compete with airlines - the Shinkansen in Japan shows that whilst immensely expensive to construct, it can be a viable alternative. By 2017, over 10 billion people have used the Shinkansen network, and the economic impact of the system has been estimated at ¥500 billion per year. On journeys under 750km, the Shinkansen has a higher market share than air travel, and is competitive up to distances of around 900-100km.