Season 2 / Breaking into the industry

Episode 2 - Diving into the business units

Romain PayetWhile Adrien spoke to the experts, I started researching the industry, largely in my bedroom. The aim was simple: understand the business units and general economics of the railway industry, so we could establish our own economic model. So I started by trying to recreate the balance sheet of a typical sleeper-train business, but working out the various costs involved. Then we picked it apart. And finally, we compared it with our potential revenue, calculated based on a market survey of both the aviation industry and existing sleeper trains.

Let’s start with the costs and the analysis we undertook. The first, and most obvious, is that relating to access to infrastructure. To put it simply, this means access to railways, to railway stops and to the electricity required for the functioning of locomotives. This information is all the more easy to find because it is made public by the various groups that manage the infrastructure in Europe. But while it may be easy to pin down, it’s rather complex to model out because businesses don’t all use the same criteria. Some look at the number of passengers that get on and off the train, others on the weight of the train, others still on the kilometres travelled.

For us, aiming as we do to run lines that travel through several countries, we needed to carry out a huge amount of analysis across a huge number of potential lines. Because at this point, we didn’t know exactly what our development plan would be. What’s more, the infrastructure costs would be among the essential criteria required to figure out this plan. So we decided to create a tool called Furlong, which would help us make our decision by bringing together all infrastructure costs in ten or so European countries in a centralised database. This would allow us to see all the associated costs laid out line by line, so we could refine our analysis.

Let’s now look at the cost of rolling stock – in our case, locomotives and other carriages. This was much less straightforward than access to infrastructure. First of all, while certain articles in the press give figures showing orders, it’s not really public information. Moreover, even the sums relayed by journalists have to be taken with a pinch of salt, because each piece of equipment is pretty much unique. The cost of a self-propelled high-speed train isn’t equal to that of a set of carriages pulled by a locomotive. Even with equivalent rolling stock, the prices can vary according to the market it is destined for. For example, a self-propelled high-speed train doesn’t cost the same if it has to be capable of travelling under the Channel. Finally, most costs mentioned in the press include several options and/or maintenance contracts that last for a long period. Because the available data is pretty opaque, we can only arrive at pretty imprecise models, and keep on testing hypotheses to refine the results.

Now that you’ve come with a rough figure for the cost of rolling stock, you have to estimate the cost of funding it. This involves a so-called ‘structuring assumption’ which every new railway business must consider. Option 1: buy stock outright, which is the solution the most widely used by historic operators. Option 2: resort to leasing, which is the solution most often used by private operators, especially in freight.

For the first, the financing cost is relatively simple to estimate as it comes down to two key points: the total potential debt across all the rolling stock and the market interest rate for this sort of debt. For the second option, on the other hand, and this point is specific to the railway sector, public information remains very rare. We could only resort to advice from experts and very rough estimates from lending firms. Happily for us, we managed to win the confidence of one of them very early on in our research. Within this company, one person in particular guided us to the more realistic of our hypotheses: the ones we should really be using in our estimations.

Now we come to personnel costs, which are incredibly important in a railway business. They can be split up into two categories: costs for headquarters staff, and costs for travelling staff. The first are closely linked with the vision of company managers, the size of the firm, its org chart and governance. Clearly, it wouldn’t make much sense to apply HQ costs at SNCF to a company that aims to start with a single line. Those costs are easily modellable if you establish a clear org chart, with clearly defined roles, the experience required for those jobs and the right salaries.

As for travelling staff, the task is clearly more complex because the costs depend, obviously, on the number of staff that you want in the train, as well as their shift pattern. A daily train operated by ten staff – a locomotive driver, an onboard chef and eight staff – clearly requires ten separate recruitment processes. That figure will increase automatically because of a collective agreement, which is publicly available information, that applies to the journey time on each line. This collective agreement is the result of decades of negotiations between leaders of existing railway businesses and their employees, and it’s very complex indeed. It includes a certain number of off-days and takes into account the number of odd-days taken on weekends, the number of days that involve staying away from home and quite a few other factors.

And so those are the main costs you have to consider when it comes to the economics of a railway business. We applied this research method, carried out largely in my bedroom and combined with the advice of experts, to all the other costs which, although less important, remain essential to the definition of any economic model.

Now that we‘ve taken into account these factors and analysed them, we’ve got a rough cost base. Our near stage will involve coming up with our potential revenue, so as to work out the level of profitability we might be able to achieve. To do that, we’ll dive into the analysis of two markets. The first is obvious: the historic sleeper-train market. And the second is our real competitor, one which we want to draw travellers away from: aviation. And that will prove quite the challenge.

Midnight Trains Logo