46th Congress of the European Regional Science Association (ERSA)
August 30th - September 3rd, 2006, Volos (Greece)
Spatial and individual influence on commuting behaviour in Germany
Georgios Papanikolaou
University of Dortmund, Faculty of Spatial Planning, Department of Economics
Dortmund, Germany, [email protected]
Abstract
In regional planning literature it is often argued that regional deconcentration process and urban sprawl tend to
increase the amount and distance of commuting (Vandersmissen et al. 2003). Supported by improved traffic
infrastructure and falling transport costs, the spatial division of labour and housing locations extends the travel-
to-work areas and compensates for locational disadvantages associated with local labour markets (Moss et al.
2004).
On the other hand, the contrary hypothesis is postulated, too. Since both working and living places are
deconcentrated, shorter commuting distances may occur because centre-orientated connections are substituted
with intra-peripheral commuting patterns (Gordon; Richardson 1991: 419).
The intention of this paper is to analyse the influence of spatial structure on commuting behaviour of employees.
For example does commuting behaviour differ between types of regions and if so what are the reasons for these
differences?
There are several ways in which the locational conditions of the living area may influence the commuting
behaviour. Firstly, the more jobs that are available, the less is c.p. the necessity for long-distance commuting.
Thus, the lower the regional density of jobs the more frequent and longer is the commuting that should be
expected. In addition, the variety of local jobs should have an influence on the commuting behaviour. As in rural
regions less varied jobs are available, there should be a pressure on the local residents to increase their job
opportunities by extending their commuting distance.
However, apart from regional influences, individual characteristics may effect commuting behaviour, too. For
example, highly qualified employees are likely to commute longer distances since they could be more interested
in realising their investment in human capital. Conversely, long-distance commuting should be c.p. less
attractive for part-time employees because of the less favourable relation between work and commuting time.
The empirical analysis consists of two parts: Firstly, regional differences in commuting behaviour are analysed.
The investigation focuses on regions and the relationship between the regional characteristics of living places
and the regional amount of commuting.
The second part includes individual influences. As they can only be observed at the level of individuals, the
focus of investigation changes from regions to individuals: In addition to regional effects the commuting
behaviour of individuals is explained by individual characteristics such as education or occupation.
The data base is given by the social security statistics. It covers all employees in the social security system and
contains information about each individual. As data on both the living and the working places are available, the
statistics provide a much greater regional differentiation than frequently used surveys do. The other information
relates to some personal characteristics (like sex or education) as well as to occupation (such as full-/part-time or
the sector of employment).
Keywords: Commuting behaviour; spatial structure, individual characteristics, employment density, Germany