Although France has one of the largest railway networks in Europe, travelling by train represents only 11% of domestic passenger travel (CGDD, 2021). Furthermore, the usage of the French railway network is very heterogeneous. While a third of the network accommodates 71% of the transportation supply (in train-km), almost half of it (45%) is dedicated to only 9% of the supply. The French national railways company, the SNCF, has until recently benefitted from a quasi-monopolistic power by operating most of the network. However, complying with the European guidelines, the French government has gradually opened domestic rail passenger travel to competition, starting from 2019 and until 2023. French regions are seizing this opportunity to challenge the SNCF with other operators, and to question the relevance of rail services for small passenger lines (“lignes de desserte fine du territoire” in French). These lines face very limited ridership for comparatively often very high operating costs. In periurban and rural areas, where most of these lines are located, private cars are the preferred mode of travel, as they may prove more convenient than a low-speed train service with only a few round-trips per day. Coach services could therefore provide a more affordable alternative to train services. On the other hand, some regions have considered an opposite approach by improving service frequency on these small lines in an attempt to boost ridership and avoid closing those lines, in fear of isolating the rural areas served by these lines. This work intends to shed light on this debate by investigating the relevance of several alternative scenarios for the provision of regional passenger mobility, which vary according to the mode used – either train or coach – and the level of service frequency. The scenarios are evaluated using a cost-benefit analysis (CBA) evaluation framework based on the French national guidelines (Quinet, 2019), in order to capture the impact on consumer surplus, operating costs, and the environmental and safety externalities (Boardman, 2006). Relatively to the existing literature on the topic (see Tirachini et al., 2010 for a review), our evaluation framework includes a very detailed cost structure for train services. This allows us to take into account economies of scale related to infrastructure costs but also operating costs when improving service frequency, with a level of detail not present in previous studies. The methodology is applied to 8 railway lines in the Région Sud Provence Alpes Côtes d’Azur in French, which features a mixture of busy and small railway lines. Four scenarios are considered: 1) a “status quo” scenario with the current rail service, 2) a “coach” scenario where the service is operated by coaches instead of trains, 3) an “improved train” scenario with improved service frequency and finally 4) an “improved coach” scenario which mimics scenario 3) but again using coaches instead of trains. As the Région Sud faces chronic issues regarding regional passenger transportation, it provides an excellent case study for this analysis. Preliminary results show that, as expected, the choice between train and coach services results from a trade-off between the lower operating costs of coach services and the greater value for users of train services resulting from a better comfort for the latter mode. The environmental impact of switching from train to coaches depends on the load factor. Although it is usually in favor of coaches, it remains extremely limited compared to operating costs and private costs with the current valuation of pollutant emissions. Improving service frequency, both for trains and cars, often proves an effective policy, with a positive net present value for seven out of eight lines. We find that the critical level of ridership between coach and train services is around 700,000 passenger-km per year for regular levels of service frequency, and around 500,000 passenger-km per year when service frequency is improved and optimized, thanks to economies of scale.