Benefit-Cost Comparisons for Rail Infrastructure: How Realistic are the Assumptions?

This text is the English translation of a LinkedIn article published in German a few days ago. The examples used are from Germany; nevertheless, they illustrate facts that apply similarly in other (European) countries. All sources and references are listed at the end of the text.

The benefit-cost comparison (BCC) is calculated for each measure in advance of publicly financed road infrastructure. In the methods manual for the German Federal Transport Infrastructure Plan 2030 (Bundesverkehrswegeplan 2030, BVWPl) (1) this procedure is described, the actual document itself (2) contains the results of this planning.? In benefit-cost calculations, all monetizable effects are recorded and the monetized benefits are compared with the costs. Only if the BCC is greater than 1 is the expected benefit greater than the costs and only then are the corresponding projects economically viable.

Benefit-cost comparisons and their assumptions

As the BCCs are carried out way ahead of the planning and even more the construction project itself, the decisive factor for the result of these analyses is which assumptions are made on both the benefit and cost side and, above all, how realistic these assumptions are. I have already explained in my article “Mobility: What does it cost? Who benefits?” that the costs of almost all recent rail infrastructure projects have been dramatically underestimated and that the actual costs after completion of the projects were many times higher than planned.

In the following, we will look at the quality of the assumptions on the benefits side. In principle, aspects such as exhaust emissions, traffic safety, travel time in passenger transport, freight transport, reliability, benefits from shifted traffic from road or air to rail, etc., are included in the benefit considerations of the BCC. For the first time, an "implicit benefit" is also included in the BCC in the BVWPl 2030, whereby it was explained in a documentation of the scientific service of the German parliament 2022 that the considerations on which the implicit benefit is based are neither recorded nor measurable (3). The question of how the aspect of reliability should be monetized in the run-up to the planning of a construction project, i.e. decades in advance, and incorporated into a BCC, can also be considered as open. Did anyone foresee decades ago that more than a third of long-distance trains would be significantly delayed in Germany in 2023 and factor this unreliability into the BCC as a negative benefit? It is obvious that such "soft" factors can be used to influence a benefit-cost comparison quite easily in order to push it in a certain direction.

However, the assumed traffic figures for both freight and passenger transport are very important for these analyses. They are the most important lever for moving BCCs in one direction or another. It is almost always about transferring traffic from road or air to rail, which is used to justify any infrastructure measures in the rail network.?

Regrettably, however, there is no detailed information in Germany on passenger kilometers travelled on individual routes in previous years, nor are concrete forecasts published for specific route sections. Much more transparency would be required here, but so far there has been no political will at all to persuade Deutsche Bahn to be more open. However, both Deutsche Bahn and the Federal Ministry of Transport (BMDV) have rejected specific inquiries about both past figures and forecasts, citing confidentiality and reasons of competition protection. It remains to be seen who Deutsche Bahn is competing with in long-distance rail transport.

In aviation, for example, the Federal Statistical Office publishes figures broken down to the individual passenger on a monthly basis for all routes served, although there is certainly competition between different airlines, at least on some routes.

There have been very occasional forecasts and also "success stories" from the railroads about passenger numbers on certain sections of line, some of which we would like to take a look at.

High-speed line Cologne - Frankfurt

DB's prestige project in the 1990s was the HGS from Cologne to the Rhine-Main region. A line was built right through the Westerwald and Taunus mountains which, in addition to dozens of viaducts, also required around 47 kilometers of tunnels, as trains can only cope with much lower gradients for physical reasons than is the case for motor vehicles in road traffic. (The A3 highway, which runs parallel to the line, does not require any tunnels at all). In March 2000, when the line was still under construction, Deutsche Bahn (DB) published "Today’s Concept for Tomorrow’s Traffic" (4); an excerpt from it reads: "The forecasts for passenger volumes underline the importance. Instead of 11 to 12 million passengers, around 20 to 25 million passengers will use the service every year by 2010."

After the line had been in operation for 15 years, DB celebrated itself, albeit with completely different figures: "Huge success: high-speed line Cologne - Rhine/Main with 220 million passengers in 15 years," it said in a press release on August 1, 2017 (5).

You don't need to have studied mathematics to work out that in the first 15 years there were not 20 to 25 million, but an annual average of less than 15 million passengers per year who "used the service, 33% to 66% less than forecasted in 2000.

High-speed line Berlin - Munich

When planning the high-speed line Berlin-Munich, which at €10 billion was considerably more expensive than originally planned, it was still assumed in 2010 that around 70 freight trains would run in each direction every day in addition to the ICE passenger trains, a key assumption on the benefit side of this project; freight traffic is ultimately to be shifted from road to rail. Expert advice that the line was only partially or not at all suitable for freight trains was ignored, as the costs (which were dramatically underestimated at the time) could no longer be justified. It came as it had to: an inquiry by the Green party in the Bundestag in 2018 revealed that not a single freight train had traveled on the new line (6). What the BCC would have looked like if the project had been planned without freight traffic from the outset is pure speculation from today's perspective, but the benefit-cost index would definitely have been worse. Calculated over just one year, that is more than 50,000 freight trains that were not run and whose cargo was not shifted from road to rail.

But there are other questionable figures for this Deutsche Bahn flagship project. After one year of operation, DB reported on the results of a study it had conducted itself, stating that it had succeeded in persuading one million travelers each from the road and from the plane to change to the train (7).

As previously mentioned, in aviation every single passenger is known; we know their name, the date, the flight number and even the seat they were traveling in. And the Federal Statistical Office publishes these figures, anonymized of course, but in such detail that the exact number of passengers is known month by month for every domestic German flight route. The DeStatis database shows that passenger numbers from Berlin to Munich and back were always between 1.93 million and 1.98 million in the years 2015-2019, with the highest figure of 1,985,300 being reached in 2018 (8). This is exactly the year for which Deutsche Bahn claims to have transferred one million passengers from the plane to the train.

Why does Deutsche Bahn make such misleading publications, which are also taken up by many media without being checked? Is it to justify the fact that every passenger traveling by train on this route has to take a detour of around 100 km and race through several kilometers of otherwise unnecessary tunnels because the route was nonsensically routed via Erfurt? Is this a distraction from internal difficulties or do questionable decisions of the past need to be justified in retrospect?

Both the Cologne - Frankfurt route and the Berlin - Munich route clearly show the quality of the assumptions made in the NKV on the benefits side. Due to the lack of traffic figures, it is difficult to create the transparency for other infrastructure projects that would be urgently needed to justify the immense financial investments and the enormous CO2 emissions of the infrastructure.??

Another criterion that is included in the calculations for the BCC in advance is travel time. In the political and societal debate, as well as on the part of Deutsche Bahn, there is a persistent claim that travelers are willing to switch from plane to train as soon as the travel time by rail does not exceed four hours. There are neither scientific studies nor any empirical data to support this claim. In fact, Hartmut Mehdorn, then CEO of Deutsche Bahn, admitted in 2002 that he always flew the Berlin - Munich route because "train journeys of more than four hours are torture". In a talk show at the time, he stated his goal of offering the Berlin - Munich route in under four hours. This statement by Mehdorn was later reinterpreted to mean that all business travelers would willingly switch from plane to train if the train journey did not take longer than four hours (9).

Conclusion

Realistic cost-benefit comparisons are urgently needed to check and ensure the economic viability of investments worth billions. However, these complex procedures only make sense if realistic assumptions are used, both in terms of benefits and costs. The BCC calculations for major infrastructure projects in recent years are often between 1 and 3, often only slightly above 1, which makes them just about economically justifiable. If the costs then rise to a multiple of the original planning, this usually means nothing other than that the project is in reality not economically viable but was nevertheless realized. If the benefits were also set unrealistically high in the BCC calculation, for example because freight traffic was planned that will never be realized or passenger numbers were calculated far too high, economic billion-euro graves are created at the taxpayer's expense. The motto is then close your eyes and get on with it, combined with the certainty that those responsible will never be called to account.

In this context, it is also interesting to note that the parliamentary group of the Greens is now calling for the BCC to no longer be carried out for certain rail infrastructure projects (10). This would really be throwing the baby out with the bathwater. The fact that BCCs are often carried out with unrealistic assumptions can hardly be compensated for by no longer carrying them out at all. Rather, reality must be adequately reflected in these calculations. And this reality has economic components as well as mathematical and scientific components that simply cannot be ignored.

Sources & References

1. PTV Planung Transport Verkehr AG, et.al . Methodenhandbuch zum Bundesverkehrswegeplan 2030. [Online] 2016. [Zitat vom: 04. 01 24.] https://bmdv.bund.de/SharedDocs/DE/Anlage/G/BVWP/bvwp-methodenhandbuch.html?nn=12830 .

2. BMDV. Bundesverkehrswegeplan 2030. [Online] [Zitat vom: 04. 01 24.] https://bmdv.bund.de/SharedDocs/DE/Publikationen/G/bundesverkehrswegeplan-2030-gesamtplan.html .

3. Deutscher Bundestag - Wissenschaftliche Dienste. Kosten-Nutzen-Verh?ltnis bei Bundesverkehrswegeprojekten. [Online] 2022. [Zitat vom: 04. 01 24.] https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&uact=8&ved=2ahUKEwiLiqObrMODAxXQh_0HHQVxD8kQFnoECBgQAQ&url=https%3A%2F%2Fwww.bundestag.de%2Fresource%2Fblob%2F918590%2F09c8c5a46f13d37237d777d6a44fc217%2FWD-5-113-22-pdf-data.pdf&usg=A .

4. DB Projektbau GmbH (Hrsg.). Ein Konzept von heute für den Verkehr von morgen - Neubaustrecke K?ln - Rhein/Main. 2000.

5. Deutsche Bahn - Regionalbüro Düsseldorf. Riesenerfolg: Schnellfahrstrecke K?ln - Rhein/Main mit 220 Millionen Fahrg?sten in 15 Jahren. 1. August 2017.

6. Süddeutsche Zeitung. Kein einziger Güterzug auf der Vorzeigestrecke. [Online] 14. 02 2019. [Zitat vom: 26. 04 23.] https://www.sueddeutsche.de/wirtschaft/deutsche-bahn-schnelltrasse-berlin-muenchen-gueterverkehr-1.4328369 .

7. SYSTEM BAHN. Ein Jahr Schnellfahrstrecke Berlin - München: Reisende wechseln vom Flieger zur Bahn. [Online] 19. 01 2019. [Zitat vom: 27. 03 2023.] https://www.system-bahn.net/aktuell/ein-jahr-schnellfahrstrecke-berlin-muenchen-reisende-wechseln-vom-flieger-zur-bahn/ .

8. Statistisches Bundesamt. DeStatis-Datenbank. [Online] [Zitat vom: 14. 11 2022.] https://www.destatis.de/DE/Home/_inhalt.html .

9. Der Spiegel. Mehdorns Gest?ndnis: Lange Bahnfahrten sind eine Tortur. [Online] 2002. [Zitat vom: 29. 03 2023.] https://www.spiegel.de/reise/aktuell/mehdorns-gestaendnis-lange-bahnfahrten-sind-eine-tortur-a-219160.html .

10. Gastel, Matthias. Schienen-Infrastruktur zügiger ausbauen. Homepage von Matthias Gastel, MdB, Sprecher für Bahnpolitik der Fraktion Bündnis 90/Die Grünen. [Online] 14. 12 2022. [Zitat vom: 04. 01 2024.] https://www.matthias-gastel.de/schienen-infrastruktur-zuegiger-ausbauen/ .

?