Draft paper prepared for the International Population Conference Cape - - PDF document

Draft paper prepared for the International Population Conference Cape Town, South Africa, 29 October – 4 November 2017 The Transition to Parenthood Among Britain’s “Generation Rent”: Examining the changing role of housing tenure Valentina Tocchionia Ann Berringtonb, Daniele Vignolic, Agnese Vitalid – THIS IS A VERY PRELIMINARY DRAFT, PLEASE DO NOT QUOTE OR CIRCULATE – Abstract A positive link between homeownership and fertility is usually presumed. Nevertheless, couples’ preferences to become homeowners before having their first child has been undermined by the dramatic changes in the UK housing market over recent decades: home-ownership rates have fallen dramatically among young adults as a result of low wages, precarious employment, reductions in the availability of mortgage credit, and rising house prices. Using prospective longitudinal data from the British Household Panel Survey and the United Kingdom Household Longitudinal Survey and applying discrete-time event-history techniques, we investigate whether and how the link between housing tenure and timing of first births has changed over recent decades in UK: from 1991 to 2015, the probability of conceiving has decreased into homeownership, whereas it has increased in private renting. Increasingly, young people remain in insecure private rented accommodation even during the process of family formation and parenthood. Keywords: Britain, fertility, housing tenure, event-history analysis, panel data Acknowledgements: The research leading to these results has received support under the European Commission's 7th Framework Programme (FP7/2013-2017) under grant agreement n°312691, InGRID - Inclusive Growth Research Infrastructure Diffusion, and under the ERC Consolidator Grant EU-FER “Economic Uncertainty and Fertility in Europe”, financed by the European Research Council under the European Union’s Horizon 2020 research and innovation programme, grant agreement n°72596 (PI: Daniele Vignoli).

a University of Florence, email: v.tocchioni@disia.unifi.it b University of Southampton, email: A.Berrington@soton.ac.uk c University of Florence, email: vignoli@disia.unifi.it d University of Southampton, email: A.Vitali@soton.ac.uk

Introduction Housing markets affect fertility both directly and indirectly (Mulder, 2006b). The direct link from housing to childbearing is that couples prefer to be homeowners before having children (Feijten & Mulder, 2002; Mulder & Wagner, 2001) and thus we would expect to observe a housing transition towards homeownership prior to childbearing (Kulu & Steele, 2013). Housing markets act indirectly on fertility through their impact on the ability of young adults to become residentially independent of the parental home. In western countries including the UK, family formation whilst co-residing in the parental home is unusual – family formation usually either coincides with, or follows residential independence from parents. Thus parent adult-child co-residence delays partnership formation and hence first births (Mulder, 2006a). The uncertainty and precariousness embedded in young people’s lives and their transition to adulthood, together with reductions in the availability of mortgage credit, lack of affordable homes, and rising house prices, make private renting the only affordable solution for many young adults who decide to live independently (Rugg, 2010). Unlike in other European countries where private rented accommodation is highly regulated, tenants in the UK have very few rights, dealing with an expensive and very insecure dwelling. Moreover, the growth of private renting in young adulthood has been much more pronounced than elsewhere (Lennartz, Arundel, & Ronald, 2015). The percentage of young adults renting privately reached 48% in 2013-2014 in England, where a decade ago the percentage stopped at 21% (Coulter, 2016). At the same time, the access to affordable home within the social rented sector has become increasingly restricted to the most vulnerable, and has sharpened housing inequalities (Berrington & Stone, 2014; Rugg & Quilgars, 2015). Housing affordability and availability vary geographically due to local variations in housing costs, average wages and in the stock of social housing available in an area. To sum up, in the UK homeownership is increasingly unaffordable and many young adults are living in privately rented accommodation at later ages, and at stages in their life course when family formation typically takes place. Young parents aged 25-34 who were renting privately grew sharply from 11.1% in 2001 to 27.6% in 2011, with a consequent fall in owner-occupation (Coulter, 2016).We therefore ask the question as to whether the positive link between homeownership and fertility remains in recent years. Indeed as noted previously for France, the cost of homeownership might compete with the costs of childbearing and childrearing, favouring childbearing among renters (Courgeau & Lelièvre, 1992). In Great Britain, the role of social renting is noteworthy, given that this kinds of housing is more secure and better perceived for childrearing compared to the private rented sector (Di Salvo & Ermisch, 1997; Ineichen, 1981; Murphy, 1984; Payne & Payne,

1977). The effects of increased private renting on fertility need still to be addressed in Britain. This project investigates whether and how the link between housing tenure and the timing of first births has changed over the last decades. More specifically, our analysis addresses the following questions: Has the probability of becoming a mother whilst in private rented accommodation as

• pposed to owner occupier or social rented increased since 1991? Is this increase explained by the

socio-economic and demographic characteristics of women in the different housing tenures? How does the probability of becoming a mother differ according to recent housing transitions, e.g. the time since last move? To what extent does the relationship between housing tenure and the progression to the first birth differ according to neighbourhood characteristics? Data and methods The study is based on a sample of women generated from the eighteen waves of the British Household Panel Survey (BHPS), in combination with data from the first six waves of Understanding Society, the UK Household Longitudinal Study (UKHLS). In total our time period spans 1991-2015. Both surveys are nationally-representative surveys based on a longitudinal design, which provide an outstanding basis for the empirical analysis of the fertility-housing nexus. We consider all women aged 18-42, who are living independently of the parental home (and hence their housing tenure represents their own, rather than their parents’ tenure) and who were interviewed at least for two consecutive waves. Discrete event-history analysis is applied to exploit the transition to the first child’s conception resulting in a live birth. Woman’s age – grouped in four categories - is the baseline

• hazard. Observations start at the first interview living independently after 18, and are censored

either when the woman has a first conception, the woman is observed in her last available wave, when she returns to the parental home, or when she reaches age 44 years, whichever occurs first. Three separate models are estimated in respect to the current housing tenure, distinguishing among homeowners, private renters, and social renters, where observations are weighted. Then, the key explanatory variable is the calendar period, distinguishing three periods: 1991-1999, 2000-2007, 2008-2015. Socio-economic and demographic characteristics including the woman’s partnership status, educational qualifications, economic activity are controlled for. By including a categorical variable for the time since move in the current dwelling, we account for anticipation strategies, scrutinizing how the hazard rate may differ as a consequence of a move. Finally, a multilevel version of the model takes account of the hierarchical structure of the data with households being clustered within primary sampling units, which are then clustered within local authority districts.

Including time varying contextual information on employment and housing markets allows us to account for the influence of the neighbourhood characteristics and hence to explore how the context shapes the hazard to have a first conception in the different types of housings. Results Figure 1 shows the housing tenures of young adults aged 25-34 who are living independently in the different family types in 1991 and in 2009-2010. During these two decades, the proportion of young adults who are renting in their late twenties and early thirties has increased among all groups, especially those without children. But even among young couples with children and young single parents the proportion has increased, suggesting that the transition to parenthood is increasingly taking place in private rented accommodation. Figure 1: Housing tenure of independent1 young adults aged 25-34 according to current family

• type. 1991 (wave 1 of BHPS) and 2009/10 (wave 1 of UKHLS).

1 Independent refers to residential independence from the parental home.

Source: Wave 1 (1991) of British Household Panel Survey and wave 1 (2009/10) of United Kingdom Household Panel Survey

Given this premises, Figure 2 shows the predicted probability from the hazard models of conceiving a first child according to housing tenure and calendar period. For both homeowners and private renters, predicted probabilities decrease in 2000-2007, whereas they increased after 2008. Noteworthy, the predicted probability in the last period remains lower that in the first period in

• wnership, instead, it is higher for private renting, thus showing an increase in giving birth in

private rented accommodation over time. On the opposite, for social renters predicted probabilities increase in 2000-2007, whereas after 2008 they decrease. Even after controlling for all confounders, the conclusions are the same (see Figure 3), thus showing how the proportion of young adults who

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1991 2009/10 1991 2009/10 1991 2009/10 1991 2009/10 Couple no child Couple with child Unpartnered no child Unpartnered with child (lone parent) Owner occupier Social renting Private renting

are conceiving a first child as private tenants has increased over the last 25 years, net of all confounders, whereas the opposite holds for homeowners.

Figure 2: Predicted probabilities of conceiving a first child according to calendar period and housing tenure. 1991-2015 Homeownership Private renting Social renting

Controlling for age group. Source: Wave 1-18 of British Household Panel Survey and wave 1-6 of United Kingdom Household Panel Survey

Figure 3: Predicted probabilities of conceiving a first child according to calendar period and housing tenure. 1991-2015 Homeownership Private renting Social renting

Controlling for age group, partnership, education, parental social class, economic activity, equivalised income (in quintiles), region, overcrowding, country of birth outside UK. Source: Wave 1-18 of British Household Panel Survey and wave 1-6 of United Kingdom Household Panel Survey

Finally, introducing a covariate for the time since move into the models and accounting for anticipation strategies, predicted probabilities of conceiving a first child follow the same path for homeowners over time, showing an increase as time since move increases, with a fall three years or more after move (see Figure 4). Instead, for private renters predicted probabilities change over time: while in 1991-1999 any meaningful difference emerges with the number of years since move, after 2000 predicted probabilities are similar to the trend of homeowners, with an increase during the first two years since move, and a loss after three years or more since move. For social renters, no trend is emerging over time, thus showing that they are a very selected group.

Figure 4: Predicted probabilities of conceiving a first child according to time since move, calendar period, and housing tenure. 1991-2015 Homeownership Private renting Social renting

Controlling for age group, partnership, education, parental social class, economic activity, equivalised income (in quintiles), region, overcrowding, country of birth outside UK. Source: Wave 1-18 of British Household Panel Survey and wave 1-6 of United Kingdom Household Panel Survey

Abridged conclusion In the twenty-first century, private rented accommodation has changed its nature and become a type

• f dwelling where couples may start a family and have the first child, if they cannot afford to

become homeowners. The findings suggest that private renters are nowadays moving home within this tenure in anticipation of becoming a parent.

References Berrington, A. M., & Stone, J. (2014). Young adults’ transitions to residential independence in Britain: the role of social and housing policy, 1–28. Retrieved from http://eprints.soton.ac.uk/365137/ Coulter, R. (2016). Social Disparities in Private Renting Amongst Young Families in England and Wales, 2001-2011. Housing, Theory and Society, 6096(November), 1–26. https://doi.org/10.1080/14036096.2016.1242511 Courgeau, D., & Lelièvre, E. (1992). Interrelations between first home-ownership, constitution of the family, and professional occupation in France. Demographic Applications of Event History Analysis, 120–140. Di Salvo, P., & Ermisch, J. (1997). Analysis of the dynamics of housing tenure choice in Britain. Journal of Urban Economics, 42(1), 1–17. Feijten, P., & Mulder, C. H. (2002). The timing of household events and housing events in the Netherlands: A longitudinal perspective. Housing Studies, 17(5), 773–792. Ineichen, B. (1981). The housing decisions of young people. British Journal of Sociology, 252–258. Kulu, H., & Steele, F. (2013). Interrelationships Between Childbearing and Housing Transitions in the Family Life Course. Demography, 50(5), 1687–1714. https://doi.org/10.1007/s13524-013- 0216-2 Lennartz, C., Arundel, R., & Ronald, R. (2015). Younger adults and homeownership in Europe through the global financial crisis. Population, Space and Place, Advance online publication. https://doi.org/10.1002/psp.1961 Mulder, C. H. (2006a). Home-ownership and family formation. Journal of Housing and the Built Environment, 21(3), 281–298. https://doi.org/10.1007/s10901-006-9050-9 Mulder, C. H. (2006b). Reflexion Population and housing : A two-sided relationship Clara H . Mulder. Mulder, C. H., & Wagner, M. (2001). The connections between family formation and first-time home ownership in the context of West Germany and the Netherlands. European Journal of Population, 17(2), 137–164. https://doi.org/10.1023/A:1010706308868 Murphy, M. J. (1984). The influence of fertility, early housing-career, and socioeconomic factors on tenure determination in contemporary Britain. Environment and Planning A, 16(10), 1303– 1318. Payne, J., & Payne, G. (1977). Housing pathways and stratification: a study of life chances in the

housing market. Journal of Social Policy, 6(2), 129–156. Rugg, J. J. (2010). Young people and housing: the need for a new policy agenda. York: Joseph Rowntree Foundation. Rugg, J. J., & Quilgars, D. J. (2015). Young People and Housing: A Review of the Present Policy and Practice Landscape. Youth and Policy, 114, 5–16.