## The Distance Factor in Swedish Bus Contracts: How Far Are Operators Willing to Go?

### Authors

Andreas Vigren, VTI

### Description

This project uses a comprehensive and unique dataset for Swedish bus tenders to investigate what impact the location (distance from depot) of operator has on their probability to participate and win tenders, as well as the impact on the placed bids.

### Abstract

The recent public transport literature has given much attention to costs and governance questions, but perhaps not as much matters regarding competition. Competitive tendering is getting mature in Europe, and not the least in Sweden where the awarding method has been around since the 90’s. Results from a related recent study (to be published) for Swedish bus tenders showed that local competition seem to matte for the number of bids a public transport authority (PTA) receive in a tender, not necessarily competition on national level. One plausible interpretation would be that the locality of the operators (for example where a depot is situated) is crucial for whether an operator decides to participate in a tender or not, which, if true, would have important policy implications. If there are areas where only a single operator has a depot, a monopoly situation might arise. In order to stimulate competition, the PTA could decide to build (or rent) a depot, which the winning operator can dispose.

For this project, a dataset with detailed information about the currently active Swedish bus contracts is used along with detailed company information containing workplace information (for example depots). Because the bidders’ identities are known, the data on the locality of the workplaces can be mapped, and distances to the lines to be run in a contract and an operator’s depot measured. The data is comprehensive, and unique in both a Swedish and European context and comprises over 500 contracts from 100 tenders spanning the years 2007-2016. The detailed data allows to control for contract and spatial factors, thus isolating well the distance effects.

For the analysis, a probit discrete choice model is used for analyzing operator participation and probability of placing winning bids, where the variable of interest is the operators’ distance to the lines in the tendered contracts. The models will give quantitative measures on the probability of participation and wins for given distances from the contract area. Thus, the magnitude and importance of locality can be inferred. In an additional analysis, linear regression will be used to see what, if any, effect distance has on the bid the operator eventually places, given that it participates.

Some (very) preliminary results indicate that distance has a substantial effect on the operators’ decision to participate or not. Having on average 10-15 km to drive from the depot to the lines in a contract can reduce the probability to participate from 60 to 20 percent. If this is the case, policy makers could consider providing depots for the winning operators. This would reduce entry barriers and could stimulate competition.

#### Publisher

Association for European Transport