Three essays on transport CBA uncertainty

Cost Benefit Analysis (CBA) has for a long time been used in transport planning, but it is often questioned. One main argument against CBA is that the results depend largely on assumptions regarding one or a few input factors, as for example the future fuel price or valuation of CO2 emissions. The three papers included in this thesis investigate some aspects of uncertainty in transport CBA calculations. The two first papers explore how changes in input data assumptions affect the CBA ranking of six rail and road investments in Stockholm. The first paper deals with the effect of different land-use assumptions while the second deals with the influence of economic growth, driving cost and public transport fare. The third paper investigates how alternative formulations of the public transport mode choice and route choice affect travel flows, ticket revenues and consumer surplus. These are important factors previously known to affect CBA results. The findings of the first two papers suggest that CBA results are robust concerning different land-use scenarios and single input factors. No change in rank between a road and a rail object is observed in the performed model calculations, and only one change between two road objects. The fact that CBA results seem robust regarding input assumptions supports the use CBA as a tool for selecting transport investments. The results in the third paper indicate that if there is detailed interest in, for example, number of boardings and ticket income from a certain transit line, or the total benefit of a price change, a more detailed formulation of the public transport mode choice and route choice will provide more reliable results. On the other hand, this formulation requires substantially more data on the transit line and price structure than the conventional formulation used in Swedish transport planning, especially in areas with many different pricing systems.