Public transport congestion costs: The case of the Paris subway

The paper argues that congestion in public transportation is alike the much more studied road congestion. It sets out to produce, on the case of the Paris subway, a congestion cost curve linking the willingness to pay for non-congested travel to actual congestion levels. Congestion costs appear high. What travelers would be ready to pay to avoid congestion in the Paris subway is on average about three times the amount of their out-of-pocket payments. An 8% increase in densities experienced over the 2002–2007 period implies a welfare loss of at least 75 M€/year. Taking into account the subway congestion costs and the resulting externalities modifies significantly the optimal car-subway mix (from 8% to 24% according to a tentative computable model).

► Public transport congestion is analytically similar to car congestion. (71 signs). ► Reducing car congestion but ignoring public transport congestion is sub-optimal.(81 signs). ► We estimate a cost curve showing comfort losses as a function of subway crowding. (82 signs). ► A model combining car and subway congestion is applied to the case of Paris. (78 signs).