带我回家，乡村路：选择自动驾驶电动汽车而非私人航空

摘要

This paper investigates the potential long‐run effects of autonomous and electric vehicles, and a carbon tax, on personal domestic aviation demand in Australia. We estimate a discrete choice disutility model with two travel modes—car and air—using Australian National Visitor Survey data and Bayesian priors. We use multiplicative Fréchet errors, consistent with a constant elasticity of substitution utility function for a representative consumer of both modes. An elasticity of substitution of almost 4 replicates the observed transition to air travel as distances increase. The joint long‐run effect of five scenarios—electrification, autonomy, overnight robotaxis, a 10 km/h car‐speed increase enabled by increased safety and an AUS$200/tCO 2 e carbon tax applied to aviation—is a 43% reduction in personal domestic air passengers. A Shapley decomposition attributes the carbon tax and autonomy as the largest contributions (−15.8 and −14.1 percentage points, respectively), followed by the speed increase (−5.5 pp), electrification (−4.7 pp) and overnight robotaxis (−3.0 pp). Reductions are highest for shorter flights, so passenger km (−35%) and emissions (−38%) decline less than passenger numbers, while aircraft trips decline more (−56%).