A Review of Computable General Equilibrium Modelling for Transport Appraisal

The traditional cost-benefit analysis (CBA) for transport appraisal has two key weaknesses. Firstly, by assuming static prices in markets external to transport, externalities cannot be considered other than in an ad-hoc, case-by-case manner. This approach is only valid under restrictive assumptions regarding perfect competition and the reflection of marginal costs in prices, which may not be appropriate for large projects with wider economic impacts. The second key weakness is that a CBA does not describe the distribution of impacts. When setting policies, it can be important to know which households or industries are the recipients of economic impacts, particularly if regional equality is an intended policy outcome. Computable general equilibrium (CGE) models simulate agents in an economy reacting to price and quantity signals. By measuring welfare at the household level, CGE models are able to account for all the market linkages that affect income. We review existing CGE models applied to transport projects and policies, broadly separated into regional or urban models. Regional models incorporate transport as a mark-up industry to production or as a reduction in the value of goods in proportion to the transport cost. Urban models extend the regional concept by using discrete choice models for residence, work and shopping decisions. However, due to the large number of equations, CGE models are data hungry and computationally intensive. The CBA retains an advantage in modelling detail as the use of a generalised cost function can incorporate all the characteristics of transport that a consumer might consider relevant.