What Determines Decision‐Makers’ Preferences for Road Investments? Evidence from the Norwegian Road Sector

Abstract

What determines decision‐makers’ preferences for road projects has been a subject of debate in the transport economics literature for decades. Because economic assessments of road projects are conducted subject to demands by decision‐makers in almost all western European countries and the USA, it should be expected that they use these assessments in one way or another to determine the preferred portfolios of projects. This paper attempts to reveal the preference of decision‐makers with respect to road investment projects to be included in the Norwegian National Transport Plan for the period 2002–11. The decision‐makers are the Norwegian parliament members. The basis for considering each individual project for investment is the Impact Assessment sheet containing monetized and non‐monetized impacts that will accrue to society if a project is implemented. The dataset comprises a pool of 1121 independent projects, of which 184 were selected for investment. We hypothesize different models that may explain decision‐making using a multinomial logit model. The preferred model shows that most of the variables determining decisions are actually included in the traditional benefit–cost analyses (BCAs), except that the decision‐maker takes account of them in non‐monetary units rather than in a composite benefit–cost ratio or net present value. Further, among the government’s three stated objectives of efficiency, safety and regional development, only safety is found to be significant in the preferred model. These results support other previous studies to the extent that a BCA per se does not matter in decision‐making, but its components matter in a non‐monetized form.

Acknowledgements

The author thanks Dr Farideh Ramjerdi, Dr Rune Elvik and Dr Lasse Fridstrøm at the Norwegian Institute of Transport Economics and Professor Jan‐Eric Nilsson at the Swedish National Road and Transport Research Institute for useful comments on an earlier version of this paper. Thanks are also due to the NPRA for supplying the data used in this study. The analysis and interpretation are the responsibility of the author and not necessarily in accord with those of the NPRA.

Notes

1. This assertion is more or less derived from previous studies that have shown that net socioeconomic impacts matter less in the ranking of road projects (see e.g. Fridstrøm and Elvik, 1997).

2. Note that the investment cost in the naïve model is included in the benefit–cost ratio.

3. Note now that we consider the decision‐makers (the parliament) to be a single decision‐maker, since they make decisions together and arrive at one decision concerning each project.

4. As suggested by one referee, the equivalent p‐values should be reported to accommodate all readers. However, due to space, a choice has to be made whether to report one or the other. Henceforth, significance levels of 5% and 10% imply p‐values of less than 0.05 and 0.1, respectively.