Survival and Exits in Neighbourhoods: A Long-Term Analyses

Abstract

Neighbourhoods form a frame for our lives. At the same time, neighbourhoods are themselves formed by mobility into and out of them. This paper studies who stays in and who leaves in two districts of Oslo. The empirical analysis is based on a survival model, estimated on a 10-year long longitudinal data-set, because neither theory nor prior studies yield sufficient guidance to build an empirical model. We propose a way to nest and test survival models and utilise this in the model specification. We find that the intensity of the outflow of native Norwegian from an area is not to any substantial degree related to the size of the immigrant population. Hence, our results do not confirm the widespread narrative of white flight as a response to an increased immigrant population in areas of Oslo. Instead, the larger part of the outflow is explained by variables related to the life-course of families. Results do not suggest that increasing the ethnic or income diversity of Oslo neighbourhoods would substantially increase outflows of native Norwegians.

Keywords::

• Neighbourhoods
• residential mobility
• migration
• survival models
• Oslo

Acknowledgements

This article has benefitted tremendously from advice and suggestions from the Editors of the Special Issue, and from the referees of Housing Studies.

Notes

¹ The terms neighbourhood, census tract and tract will be used interchangeably throughout the paper.

² Census tracts in Oslo have on average about one-fifth of the number of inhabitants that the tracts Denton & Massey (1991) refer to, have. The similarity lies in the tracts being available units for which we have statistical information. We use them as approximations for neighbourhoods.

³ Rabe-Hesketh & Skrondal (2008) describes a ρ-value of 0.1 as suggesting that there is not much dependence (p. 362).

⁴ As we do not have access to the appropriate hardware for exercising of common sense, we rely more on statistical testing.

⁵ Detailed estimation results are available upon request.

Additional information

Funding

The article is based on research funded by the Norwegian Research Council [grant number 217210/H2].