Advanced continuous-discrete model for joint time-use expenditure and mode choice estimation

•First joint time-use expenditure and mode choice model estimation with expenditure equation.•First estimation of the joint model in panel setting.•First representative VTAT estimates with standard deviations (based on the joint estimation).•First decomposition of user-type effects of the VTTS into effects of the VoL and of the VTAT.•Confirmation of the importance of expenditure information for the estimation of values of time.•First estimation of the joint time-use expenditure and mode choice model with weekly mode choice data This paper presents the joint time-use, expenditure and mode choice model, based on the theoretical framework of Jara-Díaz and Guevara (2003), for the first time estimated in panel setting while using surveyed expenditure data. This extended estimation takes into account multiple trips per individual, as well as mode availability. The model was estimated using the novel dataset gathered in Austria in 2015. It includes individual-specific information on time-use, expenditures and mode choice. As a result, we calculate the value of leisure (VoL), travel time savings (VTTS) and time assigned to travel (VTAT), that are relevant inputs to appraisals of transport policies. We also show that, at least for the Austrian working population, the omission of expenditures in the model might result in a significant overestimation of the value of leisure (16.83%); the VoL (9.29€/h) was estimated to be considerably lower than the wage rate (12.14€/h) and the VTTS varies strongly between the modes (9.98€/h for car, 3.91€/h for public transport, 9.25€/h for bike and 17.53€/h for walk). The joint estimation framework produced positive estimates of VTAT (5.38€/h) only for public transport, reflecting the favorable public transport conditions in Austria.