![Sample our Health and Social Care journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5938/TOC-Subject-Sample-2021-Health-Social-Care.jpg"})
ABSTRACT
This paper examines the social construction of house size expectations in two national panel datasets: German Socio Economic Panel Study (GSOEP) and the British Household Panel Study (BHPS). More specifically, it tests the aspiration spiral theory and positional good theory using data on housing/life satisfaction and house size judgements. In both countries, it finds substantial evidence that the current space expectations of individuals who have ‘upsized’ depends on the level of living space they experienced in the past year. For downsizers, however, the evidence in support of the aspiration spiral theory is weaker. In terms of the positional good theory, this paper finds no consistent evidence that an individual’s space expectations are influenced by those around them. In both countries, the paper tests for two reference groups – the average level of living space in the region, and the mean size of the largest decile of houses in the region – and neither are found to be significant.
Acknowledgements
This is the final paper to be published from my PhD so I’d like to thank again my supervisors, David Clapham and Tommaso Gabrieli, for all their guidance and inspiration. I am also grateful for comments from Ken Gibb and Geoff Meen (my PhD examiners), participants at the ENHR Housing Economics workshop in Aberystwyth, and the various anonymous referees.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 We use the median metres per person for East and West Germany but the mean rooms per person for the UK (as the median only changes twice over this time period, making it difficult to discern any pattern).
2 Thanks to Referee 2 for suggesting we distinguish between upsizer and downsizers.
3 From 2008 onwards the BHPS sample was subsumed into the (larger) Understanding Society sample, but unfortunately many of the housing variables were dropped in the process, making Understanding Society relatively useless for the purposes of this paper.
4 Life satisfaction has been used as a direct proxy for utility by a host of empirical studies (e.g. Clark, Citation2003; Clark and Georgellis, Citation2013).
5 In both the BHPS and GSOEP, the space shortage question was only asked in household questionnaire (as opposed to the individual questionnaire). This poses something of a dilemma later on when selecting adopting ‘reported space shortage’ as a dependent variable. On one hand, it makes sense to limit the sample to those respondents in the household who were the ‘main contributors’ to the household questionnaire (this is recorded by the interviewer). On the other hand, it maybe that multiple respondents answered the questionnaire, or that the main contributor answered on behalf of the household as a whole, suggesting we should include all members of the household. We ran the ‘space shortage’ regressions using both sample selection criteria but there were no meaningful differences in the findings. To ensure consistency, we therefore opt for the larger sample size and include all adult household members in our analysis.
6 Ideally, we would adjust the space per person measure to account for local norms around space usage. For example, in some regions in might be more socially acceptable to have children sharing a bedroom than others, thus implying a lower space requirement. In practice, however, it is impractical to accurately compute such a number.
7 Rooms per person is obviously a less accurate indicator of space than metres per person but in the GSOEP the two measures are strongly correlated (coefficient = 0.79) and separate analysis of the 2008 English Housing Survey indicate that the bedrooms per person and metres per person are also strongly correlated (coefficient = 0.77).
8 We use the mean because it includes all of the upper decile distribution. If, for example, only the size of the top one percent of largest houses increased then the mean would capture this but the median would not.
9 Note though that Bellet was able to estimate this figure at the county-level, a much smaller spatial scale
10 I merge Saarland with Rhineland; Bremen with Lower Saxony; and Hamburg with Schweslig Holstein
11 I exclude houses over 20 rooms from these calculations to avoid outliers that result from survey error
12 As a robustness check, we also re-run the BHPS-regressions using the median regional metres per person and the results were broadly similar.
13 In Germany and the UK, where housing is overwhelmingly provided by the market, there is no obvious reason why the positional good theory would not apply to some extent across all tenures. Even social renters – for whom house size will not necessarily be positively related to relative wealth – are still likely to want a ‘normal’ size of living space, and their house size is still likely to contribute to what is seen as ‘normal’ level of living space.
14 In BHPS, we exclude the Former European Community Household Panel survey low-income sub-sample from 1997 to 2001. In the GSOEP, we exclude: 1984, 1994, 2013, and 2015 migration (over-)sample; 2002 high income sample; 2010 and 2011 ‘family types’ booster
15 To calculate the effect of lag living space on upsizers in Columns 1–5, we take the sum of the coefficients on ‘lag rooms per person’ and ‘lag rooms per person#upsizer’. For downsizers, the adaptation effect simply equals the coefficient on ‘lag rooms per person’.
16 In the UK, for example, from 2008-onwards the size of all new homes has been recorded in the Energy Performance Certificate dataset.