What Goes by Decides What Follows: Dependence on Residential Relocation Decision on Previous Life Cycle and Mobility Events
A T Md Oakil, D Ettema, Utrecht University; T Arentze, H Timmermans, Eindhoven University of Technology, NL
This paper is a longitudinal analysis of residential relocation to explore the relationship between residential move and events in household and mobility issues like car ownership including the history of residential relocation.
It has been shown that long-term decisions such as car ownership, residential location, work location etc. are interdependent (Pinjari et al. 2007; Waddell et al. 2007). In addition, longer-term mobility decisions have been shown to depend on life-cycle events, such as marriage, childbirth, or divorce (Mulder and Wagner 2001; Verhoeven et al. 2005; Prillwitz and Lanzendorf 2006; Oakil et al. 2011). In particular, such decisions typically take place in response to stressors, resulting from changed household needs or aspirations. An important notion is that stressors can arise from different life cycle events and can be addressed by different longer-term decisions. For instance, a change in job location may increase commute distance (stressor), leading to a need to reduce travel time. This can be achieved by several actions, such as changing to a residential location close to work or owning a car or even combining both.
To the extent that various longer-term decisions have been analyzed, most studies have used a cross-sectional approach (e.g., Bina and Kockelman 2006; Waddell et al. 2006; Pinjari et al. 2007). Moreover, most discrete choice frameworks applied in current land use and transportation models (Ettema and Timmermans 2006; Ettema et al. 2007) ignore the history dependency that exists in longer-term mobility decision making, except few recent analysis of consideration of previous location (eg. Chen et al. 2009). People decide not only based on their current state but also depending on what has happened. This is necessary since changes in residence, work status or car ownership requires substantial investment of time and money and therefore occurs infrequently. As a consequence, the history of such changes will have a strong impact on the probability of changing house, job, or car in a given year.
This paper focuses on residential relocation decisions from the above perspective. In particular, it will be investigated how residential relocation decisions depend on the current state of the household (socio-economic status, current residential and job location etc.) and on the history of life cycle events and mobility changes.
This paper will conduct a longitudinal analysis of residential relocation dependent on life cycle and long-term mobility events. To this end, retrospective data regarding long-term mobility and life cycle events has been collected along with data on prospective mobility targets in the Utrecht region of the Netherlands. Data includes information of 20 years history on household events like cohabitation, child birth, separation and child home leaving and also on mobility events such as residential relocation and employer change and retirement. Most of the respondents are highly educated, home owner and own more than one car. A panel data analysis has been used to explore the relationship between residential move and events in both household and mobility issues. Lagged effects of such events have also been considered together with the history of residential relocation and car ownership. The paper uses a panel specification of the multinomial logit model.
Association for European Transport