Transport Tax Policy Simulations and Satellite Accounting within a CGE Framework

Abstract: Dissertation in Economics to be publicly examined in Hörsal 2, Uppsala University, on Thursday, March 6, 2003 at 10:15 am for the degree of doctor of philosophy. The examination will be conducted in English. The thesis consists of four self-contained essays:Essay 1, Transportation Satellite Accounts – Measuring the Size of the Transport Industry in Sweden in 1995, develops Transportation Satellite Accounts, where an estimate of the expenditure on different in-house transportation activities is singled out from the Swedish 1995 input-output matrices of the National Accounts. The results show that for example in-house transportation amounts to 1.7% of GDP as compared to the for-hire figure of 1.9%.Essay 2, Transportation Industry Representation in CGE Modeling – Does it Matter?, addresses the question of whether the transportation industry representation has an impact on the results of tax policy simulations in CGE models. This is done by analyzing an equal yield CO2 tax policy reform in two separate models, where only the transportation industry differs. The choice of transport industry representation might be significant, as the results were lessened, magnified, or even pointed in the other direction depending on the model used.Essay 3, Road Wear and the Kilometer Charge – A Computable General Equilibrium Analysis, investigates the effects of a re-introduction of a distance-related charge in a CGE model of Sweden. The results imply that the charge leads to reduced road wear but also that it appears that most of the reduction stems from reduced trucking activity and not from substitution towards less damaging trucks. The overall effects were modest.Essay 4, The Cost of Relying on the Wrong Power – Road Wear and the Importance of the Fourth Power Rule, investigates the effect on road wear and deformation of alternatives to the Fourth Power Law (i.e. the first through fifth powers) in a CGE model of Sweden. It also investigates the cost of designing a charge according to the wrong power. The results for example imply that the cost of choosing the wrong power is relatively small. And as designing the charge according to the first power amounts to a weight-distance charge rather than an axle-weight-distance, the results reveal that there are implicit costs of implementing the former.