Tendering Railway Public Service Obligation Contracts: A Balance Sheet

This opinion piece explains the complexities and potential benefits of tendering public service obligation contracts for railway services in the EU, as mandated by the Fourth Railway Package. It argues that while the goal is to reduce costs and improve services, successful implementation hinges on careful planning and execution, which includes well-sized service packages, a gradual implementation process, defined contract conditions, and attention to rolling stock and human resources.

The Fourth Railway Package set 2023 as the date for the obligation to tender public service obligation contracts in the European Union. Exemptions, a long transition period and plain infringements of the legal obligation are delaying tenders in many member states. However, some member states already had a long tradition of such tenders (Sweden and Germany) and others are effectively starting to tender as required by Regulation 1370/2007, as in the case of France. Lessons can be learnt from the experiences in these countries.

Experience shows that there are tangible benefits from tendering these contracts. Tendering has the potential to reduce the cost of services for public transport authorities (PTAs). Competition in the market forces railway undertakings to more effectively control costs. Cost reduction can be as much as 25% but it tends to stabilise at around 15% of the cost of providing PTAs with services. As a consequence, with the same funding PTAs can improve PSO services by introducing new routes and more frequencies, etc.

However, in order to fully benefit from the potential benefits, it is necessary to properly design the tendering process and the contracting conditions. A sweet spot needs to be identified in which enough bidders are attracted to create real competition for the market while ensuring that railway undertakings have no incentive to overbid and later not be in a position to meet their obligations, as has sometimes happened in the United Kingdom and Sweden.

For the success of tenders a fundamental element is the definition of the right lots. Experience in Germany shows that large packages of services exclude newcomers and so reduce competition in the tender. However, very small packages tend to be inefficient as service providers cannot exploit economies of scale and density. As we have previously explained (Montero et al. 2022), we recommend starting with a volume between 1.5M and 3M train-kms per contract in the first period of liberalisation, and eventually increasing to larger contract sizes between 3M and 5M train-kms as the market matures. In any case, in addition to volume, the geographical distribution of services and the existence of service facilities (stations, maintenance facilities, etc.) should be taken into consideration to define the lots to be put out for tender.

Progressive tendering is also recommended. A scheduled evolution of tenders over time enables both the incumbent and newcomers to participate in all tenders at their best capacity, and PTAs to properly manage the tendering. Progressive tendering allows all stakeholders to learn from previous tenders. Finally, tendering over the years creates a market of continual tenders, while tenders concentrated in time close the market after the allocation of contracts, making it more difficult for newcomers to establish themselves in the market.

The tendering conditions are always relevant. Key features are the duration of the contract, the awarding criteria, the allocation of risk, conditions on rolling stock and conditions on human resources. Starting with the duration of the contract, 15 years is the limit initially set in Regulation 1370/2007. This is a long duration but it is less than the regular life of rolling stock, which creates risk for bidders. Shorter durations are feasible if bidders can rely on rolling stock provided by PTAs, ROSCOs, etc.

Awarding criteria set incentives for bidders. The temptation to rely exclusively on price and award the contract to the lowest bidder should be resisted. It certainly increases the chances of reducing the cost of the provision of the service, but it encourages aggressive bidding and future problems when railway undertakings cannot meet their commitments. Experiences in the United Kingdom and Sweden indicate the benefit of taking into consideration the quality of the provision of services and ensuring continuity of service for the duration of the contract.

The allocation of risk is equally important. Again, PTAs might be tempted to pass all the operating risk to bidders. However, passing all the risk to railway undertakings will certainly increase the price required by bidders, sometimes very substantially. Regarding incentives, passing risks that are beyond the control of railway undertakings, such as increases in ridership due to the evolution of the population or the economy, certainly does not discipline the behaviour of the contractor. Limiting the risks passed to railway undertakings to those that they can effectively manage will reduce the cost of the contract for PTAs and provide the right incentives.

Rolling stock is the main cost and the main risk for railway undertakings and it determines the success of the tendering procedure. Rolling stock is expensive, its life extends beyond the usual duration of contracts, and if the contract is not renewed the company will have difficulty using it in other places due to lack of interoperability. As a result, rolling stock is the main barrier to entry for newcomers. Markets for rolling stock might emerge over time, but at the moment they are not a reality in most member states. An alternative is to have rolling stock owned by the PTA, making it available for all bidders.

Defining a fair period between the awarding of the contract and the start of operations increases the possibility for newcomers to have access to rolling stock and human resources, which facilitates market entry and therefore more competitive tendering.

Finally, incumbents have to prepare for tendering. They not only have to meet massive data sharing requirements to level the playing field for the tenders. They have to create specific teams to prepare the bids, they have to change their culture and accept that they will lose some contracts and that they have to manage costs and certainly the delivery of services after winning the contract. Experience shows, however, that incumbents have a fair chance of adapting to the new scenario, win a fair share of tenders (as in Germany and France) and become more efficient and competitive thanks to competitive awarding of public service obligation contracts.

References

Montero, Finger and Serafimova (2022). Progressive tendering of regional railway services. Working paper, Robert Schuman Centre, European University Institute, 2022/61.

Perennes, P. (2020). Competitive tendering for regional trains: What can be learnt from economic literature, in Matthias Finger & Juan Montero (eds.), Handbook on Railway Regulation. Concepts and Practice. Edward Elgar, Cheltenham.