Home moves in the early years: the impact on children in UK and US

Common sense holds that moving home is disruptive. British policies on housing offer families with children some protection from instability. Poor families with young children are more reliant on private rental, and are more likely to move in the US than in the UK. We ask whether moving home is of itself 'bad' for children, and if it is, whether the proposed transition towards a freer housing market in UK will have unintended effects on children. Research indicates the importance of stressful family events, neighbourhood conditions and residential mobility among influences on the well-being of young children, particularly in poor families. But it seldom examines these dimensions simultaneously. This project will do so, within a framework which compares the US and the UK.

This project is one of the first to exploit the opportunity to compare two large samples of families who had a baby around 2000: the Fragile Families Study in US and the Millennium Cohort Study in UK. We will follow these families and where they lived up to age 5. There is a host of information about parental capabilities and circumstances which may help account for why they moved (or not) and how well their children progress. We will be able to gauge child progress in a comparable way on behavioural adjustment, verbal ability and their general health.

We will first ask how many families, in each country, move home in a child's first five years, when, how often, how far, and why. We will describe moves as resulting in better or worse housing, parental employment or neighbourhood than the situation movers left behind, and compare movers with stayers. This leads to modelling the precursors of moving and disentangling the various factors at play; including neighbourhood, housing, and family characteristics. We then turn to the behavioural, health, and cognitive outcomes for children of various sorts of moves; distinguishing those that are infrequent and result in better housing or employment for the parents, from those which are triggered by adverse events such a family break-up, loss of employment or involve a move into a less favoured neighbourhood. We will also allow for the impact on children to vary by their age at move and family income and composition. The models will estimate a simple relationship between moves and child outcome, and then adjust for family and neighbourhood factors, yielding estimates of any direct and indirect effects on child outcomes.

The models will be as comparable as possible between two countries to see if the different residential stability regimes are reflected in different child outcomes. Care will be taken to derive indicators as closely comparable as possible on such key variables as neighbourhood quality and family poverty status. The resulting derived variables will be made available for use by other researchers. The analysis will adopt statistical techniques to allow for sample weighting, potential bias due to survey dropout, missing data, and the selectivity of mobility. The MCS also has data on siblings which will be used as a robustness check in sibling fixed effects models.

The project is designed to establish how much and in what circumstances moving home can be said to harm or enhance child development, and the extent to which the greater rate of early years home moving in the US, compared to UK, may be reflected in greater difficulties for children. These findings for the first five years of the 2000s will be of particular interest as the contrast in housing stability between the two countries is likely to diminish under reforms to social housing tenancies and housing benefit currently proposed for England. The project is confined to the first six years of the children's lives as the early years have been particularly neglected in the literature. The results will, however, form a basis for future study of residential and school mobility once the children pass into school ages.

This project contributes directly to a growing body of evidence about the importance of the early years for child development, a theme which is relevant to a number of policies on health, education, social inclusion, and family welfare. Within this context, this project focuses particularly on housing policy and its relationship to residential mobility among poor families. Although moving home is a common experience for families with pre-school children, this is a topic which has received relatively little attention. The international comparison offered by this study not only permits families to be followed over time but to compare their experiences in two policy regimes, one where residential stability for young families is relatively protected, as it was in the UK in the early 2000s, with the more open, more privately rented, housing provisions in the USA. We have designed this project and its dissemination strategy specifically to inform the likely consequences for children in UK of proposed changes to social housing which are expected to result in less stable housing tenure. We will discuss the UK housing policy issues with potential users in the public and third sectors at early stages of the research design. Knowledge about which families move, where and why will also be of practical use for the provision of a number of services, many provided by local authorities. We cannot conclude whether the research will reach the conclusion that it is important for young families to avoid moves, and/or for locally based services to have better provisions for catering to movers, but we expect there to be a useful message, one way or the other.

In order to communicate our evidence and findings to a range of audiences, we have plans to use the websites, working papers and other dissemination infrastructure of the Centre for Longitudinal Studies (CLS at IoE), the Centre for Social Exclusion at LSE (CASE) and the Graduate Center of CUNY. We are planning to hold a workshop on the housing policy issues for an audience including policy makers and practitioners from central, devolved and local government and third sector in London in early 2014. This will take place under the auspices of the CASE, which has a good network to reach such an audience. In addition, we are planning two events, in New York and at CLS in London, to report findings and introduce researchers to the possibilities for comparative research on the two datasets.