CO2 emissions

eu_without_ukMany of us strive to reduce our CO2 emissions, including those of us who walk to work and the 2000+ members of the EU Cycling Group . One area where employees of the institutions can probably make progress is the CO2 emissions caused by flying. Many of us being expats and because we work for a central administration that serve the needs of (almost) an entire continent, we tend to travel more than the average European. Much of this travelling occurs via airplanes, both for professional and for private reasons. We also receive many visitors that come to meet us or to meetings that we organise in the institutions. The rule of thumb is that travelling by airplane emits, for each passenger, about as much CO2 as travelling alone in a private car.

Among alternative transportation means, the train is the only one that is relatively clean as far as CO2 is concerned. It thus makes sense to ask 2 questions:

  • Is Brussels’ geographical location ideal to minimise CO2 emissions through the more intensive use of trains, even after Brexit which will shift eastward the centre of gravity of the EU?
  • Can the train be a reasonable alternative to flying?

The answer to the first question is rather surprising: Brussels, despite being “relegated to” the western edge of the Union after the UK leaves, remains a rather central location as far as trains are concerned.

To quantify this, we have computed travelling times by train between the EU metropolitan areas larger than 1 million people in the EU (UK cities excluded) and 5 main train hubs, Brussels included. A rough indication of travel time is given by the website of the Eurail travel passes, which are being offered to thousand young Europeans since this year. A more accurate indication is given by the trainline.eu website, which ironically is operated by a British company. We have subjectively looked at travel times for 11 September, departing after 8 am, shown in the table below for 5 major train hubs:

Click image to enlarge

Besides Brussels, the 4 other hubs were selected for the following reasons:

Paris for France and Frankfurt for Germany, the two countries that host the biggest high speed rail networks in Europe (Spain is catching up); in addition, Munich because it hooks the German network to the East of the continent, Lyon because it hooks the French network to the South of the continent.

Other hubs could have been selected with probably little differences (e.g. some other German cities). Some hubs were clearly not central, so we did not compare them to Brussels (e.g. Barcelona).

The connected metropolitan areas were selected if they complied with the following 2 criteria:

  • areas of more than 1 million inhabitants to reflect the fact that we tend to travel to big cities rather than to small places.
  • areas that connect to at least one of the 5 hubs in less than 9 hours, which leaves out the East of the continent, Scandinavia, Portugal and some cities in Spain and Italy. The idea is that one would not spend more than 9 hours on a train to go somewhere (initially, we used an 8 hour ceiling – one standard working day – but eventually we raised it to 9 hours because otherwise Madrid and Rome, 2 of the largest EU capital cities, would have been left out. This also allowed Zagreb, another capital city, to join the list).

A western bias is apparent in the list of connected cities. It simply reflects the fact that the train network is rather poor in the East of the continent and Scandinavia suffers from the poor connection between Germany and Copenhagen. Nevertheless, the list of connected cities represents 89 million people out of 128 million people living in areas of more than 1 million inhabitants in the EU27 (the West has a better rail network but it is also more densely populated).

The table above shows the population of each connected area and the fastest travel time by train between these areas and the 5 selected hubs. The row at the bottom shows the cumulated travel time if 5 teams of 27 people were to undertake all 27 one-way trips at the same time starting from each one of the 5 hubs.

And the winner is! Frankfurt with 161 hours cumulated time. Interestingly, Brussels is close behind with 168 hours of cumulated time, only 4% more than Frankfurt.

One might argue that the calculation above is biased because there are more large cities in Germany but no huge metropolitan area. We thus redid the calculation with a hypothetical weighted travel time which multiplies the actual travel time by the population of the area divided by the total population in the sample (89 million people) and multiplied by 28 (so we have a time that is more or less comparable to the real time). The second calculation thus gives a bonus to connections to big cities.

Click image to enlarge

Now, the winner is Paris at 167 (fictitious) weighted cumulated hours. Here again Brussels finishes second, in a tie with Frankfurt at 172 hours.

The surprising conclusion of these calculations is that Brussels, despite Brexit, remains a very central location with respect to the train network in Europe, possibly the best location. Thus, Brussels is well located to minimise our CO2 emissions by taking the train as often as possible instead of flying. Luxembourg, from a purely geographical point of view could be an even better location (but it does not appear in this calculation since with less than 600 thousand inhabitants it does not make it to the Eurostat list of metropolitan areas). However, the slow train connections to Belgium, France (other than Paris) and Germany. Milan (closest city to Ispra) does appear as one of the connected cities but was not retained as a hub because of its lack of centrality.

With respect to the question as to whether the train can be a credible alternative to flying, the answer is yes. In the first table above, 11 areas comprising about 44 million people connect to Brussels within about 5 hours (5 hours is competitive with Ryanair flights departing from Charleroi, shuttle time taken into account). 19 areas comprising 67 million people connect to Brussels within about 8 hours.

As far as our personal trips are concerned, we can always make the choice of taking the train, even if it is a bit slower than flying. As far as our professional trips are concerned, it is more difficult. Because our per diem is calculated according to the number of hours spent on mission, prospective train travellers will quickly be accused of travelling by train in order to increase their per diem. Moreover, train tickets can be more costly than plane tickets (because kerosene is tax-free despite its huge environmental impact, a typical example of an environmentally harmful subsidy).

One way around this problem would be to introduce a fictitious carbon tax in the OBT tool that estimates the cost of an airplane ticket. Carbon price in the EU emission trading scheme is still low (around €17/ton at time when this article is being drafted) despite recent improvements. More ambitious policies have set a price of carbon at €100/ton. If the fictitious carbon tax was set to that level, many train trips would become competitive in the OBT tool and we would be allowed to select them for our missions. Maybe it is time to revise the mobility policy of the institutions. The institutions should lead by example, shouldn’t they?

Leave a Reply