1 Separation and Housing Transitions in England and Wales Júlia Mikolai and Hill Kulu, University of Liverpool Abstract This study investigates the interrelationship between partnership and residential changes in England and Wales focusing on moves related to union dissolution. The aim is to determine whether marital and non-marital separation has a long-term effect on individuals’ residential and housing trajectories. Using data from the British Household Panel Survey (BHPS) and applying multilevel competing-risks event history models we analyse the risk of a move of single, married, cohabiting, and separated men and women to different housing types. We distinguish between moves due to separation and moves of separated people. Our analysis shows that many individuals move due to separation, as expected, but the likelihood of moving is also relatively high among separated individuals. Interestingly, separated women are more likely to move to terraced houses, whereas men tend to move to flats.
29
Embed
Separation and Housing Transitions in England and Wales€¦ · England and Wales and summarise the hypotheses of this study. Separation and housing in England and Wales In England
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Separation and Housing Transitions in England and Wales
Júlia Mikolai and Hill Kulu, University of Liverpool
Abstract
This study investigates the interrelationship between partnership and residential changes in
England and Wales focusing on moves related to union dissolution. The aim is to determine
whether marital and non-marital separation has a long-term effect on individuals’ residential
and housing trajectories. Using data from the British Household Panel Survey (BHPS) and
applying multilevel competing-risks event history models we analyse the risk of a move of
single, married, cohabiting, and separated men and women to different housing types. We
distinguish between moves due to separation and moves of separated people. Our analysis
shows that many individuals move due to separation, as expected, but the likelihood of
moving is also relatively high among separated individuals. Interestingly, separated women
are more likely to move to terraced houses, whereas men tend to move to flats.
2
Introduction
This paper investigates the interrelationship between partnership and residential histories in
England and Wales focusing on moves related to union dissolution. Partnership events, such
as the start and end of a co-residential union, usually trigger a move as they imply a change
of residence for at least one of the partners (Dewilde 2008; Mulder and Wagner 1998; Mulder
2006; Clark 2013; Mulder and Lauster 2010). Moves related to union formation are usually
‘upward’ and are directed towards finding an ideal home, whereas moves related to union
dissolution are ‘downward’. In other words, following union dissolution, individuals are
likely to move to smaller, lower quality dwellings (Feijten 2005; Gober 1992) because moves
after separation are usually urgent and financially restricted (Feijten and van Ham 2007).
The link between separation and residential moves is of relevance to policy makers
and town planners because increasing divorce and union dissolution rates exacerbate the
magnitude of moves and imply an increased need for more, and smaller dwellings (Mulder
and Malmberg 2011; Dieleman and Schouw 1989; Feijten 2005). Additionally, the way in
which people from different socio-economic backgrounds move in relation to separation will
influence demand and supply on the local housing market (Mulder and Malmberg 2011). The
social stratification of access to housing and of family formation (Feijten and van Ham 2010;
Dewilde 2008) makes England and Wales an interesting and important context for the study
of the interrelationship between separation and residential relocations.
Previous research has shown that separated individuals are more likely to move than
those who are single or in a relationship, they are likely to move out of homeownership, and
from single-family to multi-family dwellings (Feijten and van Ham 2007, 2010; Mulder and
Malmberg 2011; Feijten 2005; Lersch and Vidal 2014; Sullivan 1986; Dieleman and Schouw
1989; Speare and Goldscheider 1987). Although previous studies have advanced our
understanding of housing changes related to separation, they suffer from shortcomings. First,
3
distinguishing between moves due to separation (also called induced moves or event moves)
and moves of separated people (also referred to as state moves) is essential to understand
whether and how separation influences housing careers in short and long term. Most previous
studies have either not distinguished between such moves or have excluded moves due to
separation from their analysis (except Feijten 2005). Second, previous research has not
accounted for possible selection effects (except Lersch and Vidal 2014); individuals who are
more likely to move might also be more likely to separate due to unobserved characteristics.
If this is true we would overestimate the potential effect of separation on moving. Third,
previous research has either not investigated the destination of residential moves or has solely
focused on moves out of homeownership and from single- to multi-family dwellings (Feijten
and van Ham 2010; Sullivan 1986; Lersch and Vidal 2014). However, housing type at
destination might be a better indicator of socio-economic differences in a country where over
60 per cent of the housing stock is owner occupied (Department for Communities and Local
Government 2015).
We aim to fill these gaps in the literature by, first, comparing moving risks of
separated individuals to those who are single or who are in a co-residential union in England
and Wales. We then analyse moving risks by housing type at destination. Finally, we
investigate whether the impact of separation on people’s residential and housing careers is
temporary or long-term. We distinguish between moves due to separation and moves of
separated people using time since separation. Additionally, we account for possible
unobserved selection effects into separation and residential moves.
Separation and residential moves
A change in family size or family structure usually triggers a move in order to adjust housing
conditions to the new circumstances (Mulder and Lauster 2010; Kulu 2008). According to
4
utility maximization theory, residential moves should only take place if the benefits of the
move exceed its costs. Thus, each subsequent dwelling is assumed to meet the housing needs
of a household better than the previous one (Feijten 2005). Therefore, union formation and
childbirth usually lead to an ‘upward’ move on the housing ladder.
However, moves related to separation are different because they are urgent and
financially restricted (Feijten and van Ham 2007)1. By definition, upon separation at least one
of the partners has to move out of the joint home. This suggests that individuals are likely to
settle for any type of housing (often moving to parent’s or friends’ place) even if it is of low
quality and not in the preferred area. Additionally, separation leads to a lower household
income, a decrease in resources due to loss of economies of scale, and a division of savings
and assets. Therefore, moves after separation are usually to smaller, lower quality dwellings,
from home ownership to renting, and from single-family to multiple-family dwellings (Gober
1992; Feijten 2005; Feijten and van Ham 2007).
Separation may also have a long-lasting impact on individuals’ residential careers. As
housing situation right after separation is likely to be temporary, it may take several
adjustment moves before separated individuals acquire housing of a similar quality, size, and
type as before separation (Feijten and van Ham 2007; Dieleman and Schouw 1989).
Additionally, those who remained in the matrimonial home upon separation may be unable to
pay for housing costs such as rent, mortgage and maintenance (Feijten and Mulder 2010).
Thus, separated people are expected to move more often than those in a relationship.
Previous research from the Netherlands has found that separated individuals (whether
single or re-partnered) are more likely to move than steady single and those in a first union
and that this relationship persists over time since separation, although the differences become
1 Throughout this paper, we use the term 'separation' to denote the dissolution of both marital and non-marital
co-residential unions. This is because even in case of a divorce, it is usually the actual date of separation (and
not the date of the legal divorce) which implies an immediate move out of the joint home for at least one of the
partners (Feijten 2005).
5
smaller (Feijten and van Ham 2007). Similarly, in the UK Feijten and van Ham (2010) found
that recently divorced and separated people have much higher risks of moving than those who
are in a relationship. Although their moving risk decreases over time, it remains higher than
for partnered individuals. Thus, these studies conclude that separation has a long-term effect
on housing careers. However, these studies only investigated moves of separated people and
excluded moves due to separation.
A study by Feijten (2005) found that immediately after separation (0-3 months)
individuals are most likely to move to shared dwellings followed by rental accommodation.
Additionally, for women separation had a long-term effect on the risk of moving out of
homeownership but for men it only had a temporary effect (Feijten 2005). Similarly, Feijten
and Mulder (2010) found that the long-term effects of separation are stronger for women; 80
percent of moves of men who separated two or more years ago were to single-family housing,
whereas this was only 60 percent among women. Additionally, immediately after separation
women are better off in terms of type of housing (they usually stay in the joint single-family
home especially if they have the children), whereas men are better off in terms of tenure
(owner occupied dwellings). One explanation for such gender differences in the Netherlands
is that upon divorce women are usually left with fewer resources than men because women
earn less and most women with children work part-time (or do not work). For men, a decline
in resources can occur when they have to pay alimony. Interestingly, Lersch and Vidal (2014)
found no gender differences in the risk of re-entering homeownership after separation in
Britain and in Germany.
To summarise, previous studies have shown that separated individuals have higher
risks of moving than those who are single or are in a relationship, that they are more likely to
move out of home ownership and out of single-family dwellings. Most of these studies have
typically focused on housing tenure and moves out of homeownership as an indicator of
6
socio-economic status following separation and they have not studied the destination of
moves by dwelling type2. Additionally, some studies suggest that separation has a long-term
effect on housing careers but these studies do not distinguish between moves due to
separation and moves of separated individuals. Furthermore, while some authors suggest that
the effect of separation on housing careers over time is different for men and women, others
find no gender differences. Thus, there is clearly a need to explicitly study the link between
separation and housing changes to improve our understanding of this, potentially complex,
relationship. The following sections briefly describe trends in separation and housing in
England and Wales and summarise the hypotheses of this study.
Separation and housing in England and Wales
In England and Wales the number of divorces increased from about 24,000 per year in 1960
to about 150,000 a year in the early 1980s and the figure remained fairly constant until 2003.
Between 2003 and 2009, the number of divorces declined to about 114,000 per year followed
by an increase in 2010 after which it has remained stable (Office for National Statistics
2012). At the same time, the number of marriages has dropped from 415,000 a year to
250,000 between the early 1970s and 2012 (Office for National Statistics 2012). Thus, the
fall in the number of divorces between 2003 and 2009 coincides with the decreasing number
of marriages. In line with this trend, the proportion of first unions that started as cohabitation
as opposed to marriage has increased (Beaujouan and Ní Bhrolcháin 2011). As cohabiting
unions are less stable than marriages, this increase in the prevalence of cohabitation is likely
to contribute to an increase in the number of unions that end with separation (Feijten and van
Ham 2010). Although official statistics on the prevalence of cohabitation and dissolution of
cohabitation are not available, survey estimates show that approximately 60 percent of
2 One exception is Feijten and van Ham (2010) who found that separated individuals in the UK are three times
as likely as married individuals to move to a flat or shared accommodation while those who are married mainly
move to single-family dwellings.
7
cohabitations become marriages while about 30 percent ends in union dissolution within 10
years (Ermisch and Francesconi 2000; Hannemann and Kulu 2015). The dissolution of
marital and non-marital co-residential unions may have different impact on housing careers.
Cohabiting couples are less likely to own a home together or to have invested in their housing
(Feijten and van Ham 2010). This suggests that there is less to loose when cohabitors split up
compared to the dissolution of a marriage. However, for couples who are long-term
cohabitors, the effects of separation might be similar to that of a divorce.
According to the 2011 census, there were 24 million households in England and
Wales. The housing market was dominated by owner occupied dwellings (61%), 17 percent
of the dwellings was socially rented while 16 percent was privately rented (Office for
National Statistics 2014). In England, the proportion of households in owner occupied
dwellings has increased from the 1980s to 2003 when it reached a peak of 71 percent. Since
then, there has been a gradual decline to 63 percent in 2014. The proportion of private sector
households remained at about 10 percent throughout the 1980s and 1990s. Since then, the
sector has undergone sharp growth due to e.g. the introduction of assured shorthold tenancies
and the buy-to-let mortgage. By 2014, 19 percent of the households was renting privately in
England. The proportion of households in socially rented accommodation (this includes
housing from local authorities and housing associations) has decreased from 31 percent in
1980 to 17 percent in 2014. This is primarily due to the introduction of the Right to Buy
program which enabled many tenants to purchase their homes at a discounted price
(Department for Communities and Local Government 2015). The trends were very similar in
Wales where after 1991/1992, the proportion of owner occupied and privately rented
dwellings increased while that of socially rented dwellings decreased, although the proportion
of owner occupied dwellings decreased somewhat since 2000-01 (Statistics for Wales 2015).
8
Hypotheses
Based on the theoretical arguments and in light of previous findings, we develop the
following hypotheses. First, we expect that separated men and women in England and Wales
are more likely to move than those who are single or who are in a co-residential partnership
(moves hypothesis).
Second, we expect that separated individuals will be more likely to move to smaller
dwellings such as a flat, shared accommodation, or terraced house compared to their single
counterparts and particularly to those who are in a steady relationship (destination
hypothesis). This is expected to be especially so immediately after separation (short-term
effect). As separated individuals make adjustment moves later on after separation, we expect
that separation has a long-term effect on individual’s residential careers (duration
hypothesis). We expect that individuals will be more likely to move to a detached and semi-
detached house sometime after separation than they were immediately after separation,
although it will be interesting to find out whether and how the patterns vary by population
subgroups. Finally, as a previous study in Britain found no gender differences in the effect of
separation on moves, we do not expect to find such differences (gender hypothesis). In
Britain most women with children return to the labour market after a short period of
maternity leave, which may leave them with more resources upon divorce than in the
Netherlands or Germany. Additionally, in Britain housing benefits and social housing is
available for single mothers.
Data and Methods
We use data from 18 waves of the British Household Panel Survey (BHPS), a nationally
representative sample of 5,500 households and about 10,300 individuals. Between 1991 and
2008, the same sample of adults was interviewed each year. If a household splits, the original
household members are still followed up and new household members are also interviewed.
9
In our analyses, we use information on original sample members and two additional sub-
samples (the European Community Household Panel and the Wales Extension Sample). We
exclude Scotland and Northern Ireland from the sample, as in these countries of Great Britain
social norms, behaviours, and legislation are different from that in England and Wales.
Individuals are observed from age 16 or from the entry into the study (if later) until age 50,
widowhood, or the end of observation, whichever happens first. We use a sample of 4797
men and 5019 women and prepare a person-months dataset to study the risk of moving. For
the study of residential moves, panel attrition might be an issue because individuals with high
geographic mobility are more likely to be lost to follow-up than those with low mobility
(Uhrig 2008). Rabe and Taylor (2010) and Washbrook et al. (2014) found that attrition in
BHPS is not related to mobility rates and thus it does not influence analyses related to
moving risks.
We estimate four sets of models. First, we focus on the relationship between
partnership status and moves (Model 1). Second, in order to distinguish moves due to
separation from moves of separated individuals we split the category of separated individuals
by time since separation (0-4 months after separation versus 5 or more months after
separation) (Model 2). Third, we analyse moves of separated individuals by order of move
(Model 3). Preliminary analysis showed that in the first four months following separation
most individuals only move once; we therefore study the moves that happened five or more
month after separation by order of move. Finally, we include information on previous
residential history to further investigate whether separation has a long-term effect on moving
risks (Model 4).
We apply multi-level event history models with a piecewise linear baseline hazard to
study the risk of a move by partnership status for men and women. Multi-level models are
estimated because each individual can experience several moves. These moves are not
10
independent from each other and standard errors, therefore, need to be adjusted. The risk of a