How many in a crowd? Assessing overcrowding measures in Australian housing

Abstract

Key points Given the importance of housing in shaping life’s outcomes and the significant public investment in housing assistance, it is critical that policy is guided by measures of overcrowding that meaningfully reflect housing adequacy. Very few Australians live in housing that could be considered overcrowded. The norm is for households to have spare bedrooms. Over 90 per cent of the population live in homes with at least one bedroom per couple or unpartnered occupant. Despite this, there are groups within our society prone to experiencing overcrowding, leading to a range of substantial adverse consequences. These include Aboriginal and Torres Strait Islander people, people from CALD (culturally and linguistically diverse) backgrounds, of lower socio-economic backgrounds and women escaping domestic violence situations, among others. The relationship between household occupant density and occupant wellbeing is highly nuanced. There is both a conceptual and empirical disconnect between current measures of overcrowding and actual experiences of excessive density. For many purposes of policy and practice, measures of overcrowding based on simple metrics of household composition and number of bedrooms cannot adequately discriminate between crowded and uncrowded households. This includes the commonly used Canadian National Occupancy Standard (CNOS). Qualitative and quantitative evidence reject the validity of the assumptions underlying the CNOS. Measurement challenges include the conflation of the effects of occupancy and density; differential effects within and between households; and opposing positive and negative effects of higher density. Positive impacts of living in a large, well-functioning household include caring for family members, strengthened family ties, promotion of cultural identity, and financial benefits. Negative impacts include lack of privacy, noise, antisocial behaviour, health and wellbeing concerns, and family strain. For targeted groups, overcrowding needs to be reframed away from a density measure to instead capture the subjective reaction to living in a crowded environment, as well as indicators of household functioning as a key moderator. In large scale surveys, measurement of crowding may be improved by incorporating wider measures of available space and facilities in the household, rather than focussing on the number of bedrooms; and accounting for the adverse effects of multiple family units. Key findings Key channels through which higher household density impacts upon occupants’ psychological wellbeing are through over-stimulation, loss of a sense of a locus of control and of lack of privacy. Interviews with stakeholders and householders highlighted shortcomings of the CNOS as an indicator of overcrowding due to its inability to account for cultural differences in norms around shared living and sleeping arrangements, notably for people from Indigenous backgrounds, the failure to account for the capacity of different households to manage higher occupant numbers, and whether living in a large household is an active choice. The relationships between household density and selected measures of wellbeing were explored through extensive quantitative modelling using 19 annual waves of data from the Household, Income and Labour Dynamics in Australia (HILDA) survey. It also included an additional analysis of data for humanitarian migrants from five waves of the Building a New Life in Australia (BNLA) survey. The objective of the modelling was to provide a better understanding of the levels and circumstances at which household density levels are associated with adverse impacts on wellbeing as a guide to validating and improving measures of overcrowding. We find the relationship between household density and occupant wellbeing to be highly nuanced. Often, negative associations between occupant density and wellbeing are observed at very low levels of density (that is, in uncrowded households), but this gradient flattens and even becomes positive at higher levels of density. This is contrary to theoretical expectations. Key findings contributing to the complexity of these relationships include: For the general adult population, on average there is a decline in wellbeing as the number of occupants in a household increases above two, leading to conflation of the effects of occupancy and density. The simple count of household occupants or density ratios to bedrooms generally perform better than the CNOS in explaining variation in mental health, psychological distress, physical health and people’s satisfaction with their home. This implies the assumptions embodied in the CNOS on appropriate sharing of bedrooms, conditional on age and gender, have limited validity when it comes to distinguishing between crowded and uncrowded households. Adverse associations between higher household density and wellbeing apply primarily to parents, with small and even positive associations for other adult occupants. Multiple families living in the same home has a substantial negative impact on occupants’ wellbeing in addition to any effect on household density. Following insights from the qualitative interviews, we tested for differential effects of crowding by household functioning. Even using a rudimentary proxy of family functioning, based on satisfaction with intra-family relationships, we confirm this to be a significant moderator. Well-functioning households have a greater capacity to manage higher occupant density. The strongest evidence of occupant density translating to overcrowding is obtained from estimating the effect of a higher number of occupants for homes with a fixed number of bedrooms, and vice versa. Adverse associations between occupancy and mental wellbeing are strongest in smaller (2-bedroom) homes. For the application of the CNOS as an indicator of overcrowding, this suggests extra bedrooms required should be given greater weight for smaller homes. Australians have strong preferences for more bedrooms. Models of people’s satisfaction with their home imply four or five-bedroom homes are preferred, even for households with just two or three people. Australians of Asian background live in higher density households but little support is found for the hypothesis that cultural norms leave them less sensitive to impacts of higher density. Evidence from the BNLA suggests that recent humanitarian migrants are significantly more likely to live in crowded housing, and wellbeing increases with household density beyond levels that would normally be considered as overcrowding. The experiences of people from CALD (culturally and linguistically diverse) and Indigenous backgrounds living in overcrowded households were further explored in the qualitative interviews. Three primary living arrangements identified as common among overcrowded houses were large nuclear families or extended family groupings, having visitors come to stay (particularly for Indigenous households), and house share arrangements (particularly for CALD households). Key reasons for those families and people living in overcrowded housing included lack of availability of appropriate housing, compounded by difficulties accessing public housing, and discrimination in the private rental market. Cultural norms and obligations contribute to overcrowding for some people from CALD and Indigenous backgrounds. For the latter, visitors were often related to Indigenous mobility between remote communities and regional and urban centres for the purposes of accessing services. The interviews revealed substantial negative effects associated with overcrowding, including lack of privacy, excessive noise, incidents of antisocial behaviour, child safety and wellbeing concerns, increased housework, food theft, and family and financial strain. Family strain heightened by overcrowding can lead to irrevocable relationship breakdowns and family violence. Service providers are also impacted by having to manage additional repairs and maintenance, provision of intensive tenancy support and the need to reallocate tenants. Some positive effects of larger households were noted, including caring for family members, strengthened family ties, promotion of cultural identity, companionship and financial benefits. The realisation of these benefits generally relies on the household being well-functioning. Potential measures to address overcrowding are identified at a systems and service-level. Systems-level measures primarily need to address the supply of appropriate and affordable housing, and tenant-to-property allocation processes. These include increasing the level and diversity of the housing stock, more efficient repairs and maintenance, and addressing rental market failures. At the service-level, housing providers play a key role in maximising the use of existing housing capacity, sourcing alternative accommodation, and providing or linking tenants to needed services. Greater staffing and funding would enhance the capability of housing and service providers to manage and mitigate the incidence and adverse effects of overcrowding. A number of stakeholders advocated for community-controlled housing organisations, particularly servicing Aboriginal communities, as a model to provide more culturally appropriate management of overcrowding. Policy and practice implications for measuring overcrowding A key finding of this research is that, due to the complex and nuanced relationship between household density and overcrowding, current measures of overcrowding based on readily observable objective variables, such as household composition and the number of bedrooms, will have at best a tenuous link to actual experiences of crowding.