A Nested approach to car use : from driving license to car ownership and annual mileage

Many signs suggest a transition in travel behaviors from the early 2000's. For instance, the average number of daily trips per individual is decreasing in French cities, going along with a modal shift from car to alternative transportation means (transit, walking, cycling). But also, average car travel per capita - expressed in vehicles*km - has leveled off at the national level from the early 2000's, in France like in other developed countries. A better understanding of such an evolution is required. Among the key factors, rising fuel prices might have played a decisive part, in particular the peak that was attained in 2000. Beyond temporary reactions to fuel price volatility, the sudden increase of 2000 seems to be a turning point in household behaviors, as if they had become more aware of the long-lasting rising trend in fuel prices. Aiming to explain this transition, the assumption of a long-run path towards saturation will also be considered. On the long-run, average car travel per adult grows in line with the motorization rate - defined as the proportion of adults being the main user of at least one vehicle, or said otherwise, having an autonomous access to a car. Correlatively, average car travel per motorized adult remains quite stable on the long-run. The relationship between growth in individual car ownership and growth in average car travel per adult has been resulting both from a modal shift from other transportation means towards car, along with the diffusion of car ownership, and from an ever-more individualized mobility, as some car trips that were before realized in common have been increasingly realized alone (i.e are leading to a decreasing occupancy rate of vehicles). As average car traffic generated per motorized adult remains quite stable, one can expect average car travel per adult to level off, once everyone is endowed with a car. In order to get a better understanding of the apparent transition in travel behaviors from the 2000's, it seems necessary to disentangle contributions from the underlying diffusion process of driving license holding and car ownership, and temporal variations of car travel for motorized individuals. The analysis is built up of two parts: - In the first part, the determinants for driving license holding, car ownership of driving license holders and traffic generated per motorized adult are successively studied through a nested descriptive statistical analysis, relying upon French Car Fleet Surveys, consisting of a panel of 6 - 7000 yearly volunteer respondents to a self-administered mailing survey, providing information about car ownership and car travel over a long period. From 1993, information on driving license holding for every adult of the household is also available. The study covers the period from 1994 to 2010, during which the transition in travel behaviors occurred. - The 2nd part is based on a modeling framework at individual level, using panel data econometrics. Models on the probability of holding a driving license, on the probability of being motorized for driving license holders, and finally on the average car travel generated by every motorized adult are successively estimated. Probabilities for driving license holding and car ownership, conditional on driving license holding, are modeled through a probit error component framework, a discrete choice model adapted to panel data. Average car travel per motorized adult is modeled through a (log)-linear panel data model, also based on an error-component framework. The analysis is segmented by gender. Indeed, auto-mobility- driving license, car ownership, car travel - cannot be considered as a homogeneous process for men and women. First, driving license diffusion has happened later among women, and has always remained at a lower level in the generations born before 1970. The gender gap has only gradually diminished from a generation to the following, before finally disappearing in the generations born after 1970. Second, the diffusion of car ownership as an individual item is also less advanced among women. If driving license and car ownership rates seem now close to saturation among men, they are still increasing for women. Correlatively, driving license and car ownership rates are simultaneously more heterogeneous, and more sensitive to income, but also to land use patterns, workforce participation and marital status among women. However, differences between men and women cannot be interpreted exclusively in terms of a gender gap. In particular, the definition of individual car ownership is partly conventional, as the principal - and often the only - vehicle of the household is attributed to the head of household, which is usually identified as the man. Yet, this vehicle may respond to a multipurpose and collective use, not only for the head of household. Thus, apparent differences between men and women partly cover different stages in the process of diffusion for the principal vehicle and for the 2nd vehicle of the household, also corresponding to a different status as a good: the 2nd vehicle, which is at an early stage of diffusion, is still an intermediary good, while the principal vehicle of the household has become an essential. Finally, average car travel per motorized adult is higher for men than for women (as main drivers), a result that can also be interpreted in terms of vehicle specialization. The multipurpose function of the 1st vehicle, also being used for long-distance leisure trips, could explain why its use is more sensitive to economic constraints - income as well as fuel prices. Model estimation by period reveals that individuals have become more sensitive to fuel prices after 2000. Moreover, a downward trend in average car travel per motorized adult after 2000 can be noticed, indicating a new will to control the budgetary risk of car travel that was not perceived until then. Also, there has been an acceleration of the transition from gasoline to diesel engines from then. These results suggest a shock leading to a sudden awareness of the long-run raising trend in fuel prices, rather than simple temporary short-term reactions to fuel prices, or the proximity of demand saturation.


  • English

Media Info

  • Media Type: Digital/other
  • Pagination: 20p

Subject/Index Terms

Filing Info

  • Accession Number: 01688919
  • Record Type: Publication
  • Source Agency: Institut Francais des Sciences et Technologies des Transports, de l'Amenagement et des Reseaux (IFSTTAR)
  • Files: ITRD
  • Created Date: Dec 18 2018 10:18AM