Browse

Changes in patterns of family formation in Britain in recent decades have been documented in the previous chapter. In tandem with this, there have been changes in the labour market. As Chapter Four will show, male employment has become less secure while women are entering the workforce in increasing numbers, albeit often in part-time rather than full-time jobs.

The contrast with marriage and family life in the 1950s is stark, where the economic roles of spouses were symmetrical but markedly different. Men had jobs, and in an era of full employment, this meant that virtually all men worked between the end of full-time schooling and the statutory retirement age. There is a debate as to how to characterise women’s longer-term employment histories, but with some specific exceptions, women left the paid labour force at marriage, or at the birth of their first child, or they may never even have entered it. A minority of those who left during the early part of marriage returned later. These patterns gave a characteristic lop-sided M-shape to the graph of women’s employment by age (Dale, 1987). Instead of paid work in the labour force, during the early stage of family formation women took virtually sole responsibility for unpaid work within the household, and maintained this irrespective of any subsequent re-attachment to the labour market. The evidence suggests that the total amounts of paid work done by men roughly balanced the total time spent on unpaid work by women (Young and Willmott, 1974). However, the work itself was strongly segregated by gender.

This chapter is about the changes in people’s incomes from one year to the next in Britain. It aims to establish salient facts about income dynamics in general and poverty dynamics in particular, and their socio-economic correlates, drawing on new evidence for the 1990s derived from the British Household Panel Survey.

A household’s income level is strongly associated with two main characteristics: the composition of the family and the employment of its members. So this chapter follows logically from previous contributions to this book, especially Chapters Two and Four. As will be shown later in this chapter, movements in and out of poverty are related to both demographic and economic changes.

The pattern of income changes from one year to the next is one of much mobility, but most of the changes are short-range. For example, of those who are poor in one year, almost one half are not poor the following year – but those who escape poverty often remain on low incomes and have a high risk of returning to poverty in future years. Income mobility also means that the proportion of the population that is touched by poverty over a six-year period is twice as large as the proportion that is poor in any one year.

Income and poverty dynamics have intrinsic social relevance and policy significance. The extent of mobility and poverty persistence are important social indicators to be placed alongside information about the income distribution at a point in time. For example, the former Secretary of State for Social Security Peter Lilley discounted the rising incidence of low income during the 1980s with reference to new evidence about

income mobility:

People’s housing situation changes through stages of their life-course – the way in which this occurs in Britain is the focus of this chapter. In keeping with the rest of this book, the chapter is concerned with the dynamics of the housing situation, and not simply cross-sectional snapshots. It considers not only the nature of the moves that people make from year to year, but goes beyond this to look at the cumulative impacts of these moves. It then describes housing careers as they develop over the life-course.

The housing career may be considered as not unlike the work career. It starts with leaving the childhood situation, a process of relatively active search. It then leads to relatively high mobility, and a process of investment to obtain or keep better jobs or housing, although in the case of the work career the investments are mainly in human capital, while in the housing career they are financial investments. These processes of career building in both housing and work tend to lead to a stabilisation in middle periods of the life-course. It is at the later stages of the life-course that the patterns diverge, since although there is sometimes a scaling down of housing requirements, it is nothing like as deterministic as the disengagement from the labour market.

However, there are three important reasons for elaborations to this view of the housing career. The housing career is:

• substantially shaped by the work career, both because the investments which people can make in housing will depend on their incomes (and on the stability of their incomes) and also because work career mobility will often require housing mobility;

• directly influenced by the movements through various stages of the life-course: there are changing housing space and location requirements from different stages of the process of family building;

• strongly influenced by the pattern of state intervention, perhaps even more so than the work career, leading to much greater differences in housing careers in different countries, and greater changes over time within a single country.

The album tells us stories. Perhaps the studious child, curled up with a book in the corner of the frame of an old black and white photo at the start of the album, reappears in a graduation photo towards the end. Perhaps the mother-to-be is found again, as we turn the pages, with two toddlers and a less convincing smile. The walk-up flat in the background becomes a three-bedroom semi, and later acquires a roof-light and a downstairs extension, or the semi is exchanged for a studio apartment with a care assistant down the hall. Some faces recur throughout the book, older but still recognisable; we see others for a few pages, and then no more.

Each snap tells us something, but we learn more from the sequence of photographs, and more still from the connections we make between the people shown in them…. The whole album provides a picture that is more than the sum of the individual pictures, more than we would get from, say, a random collection of photos from different families in successive decades of the century. The family album tells about the complex pattern of continuity and change that make up the lives of individuals and households. (Buck and others, 1994, p 10)

‘The family’ is a subject of enormous academic, political and popular interest. It is a central feature of most people’s lives, the framework within which other relationships, activities and events take place. Families have changed hugely during the past generation: not only in the formal demographics of marriage, cohabitation and childbearing, but also in the social and economic relationships between men and women, and between adults and children.

The BHPS data on family structures, employment, income and housing, on which previous chapters are based, have been the subject of detailed analysis ever since the panel data first came on stream in the mid-1990s. Much of the material covered so far has summarised work that has already been published in a series of more detailed, and sometimes more technically complex, papers. However, the survey also includes a substantial set of questions about respondents’ state of health, and their use of health services. These have not been analysed in anywhere near as much detail, and certainly not in a way that takes full advantage of the longitudinal structure of the data. The purpose of this chapter is to develop the analysis of the dynamics of ill-health. However, because the analysis of this part of the BHPS data is at a much earlier stage, it is necessary to start by considering some more technical issues than needed to be addressed in other chapters. The most commonly used survey-based measures of ill-health and impairment in Britain are derived from cross-sectional surveys. A sample of respondents is interviewed once, and asked questions about their current state of health. This provides an estimate of the number of people ill or impaired at any time, but it provides no direct indication of the rate at which people become ill or recover. This is true of the self-reported health measures obtained by, for example, the General Household Survey (ONS, 2000) and the 1996 Health Survey for England (Prescott-Clarke and Primatesta, 1998); and of the impairment measures obtained by the 1985 Disability Survey (Martin and others, 1988), the 1995 Health Survey for England (Prescott-Clarke and Primatesta, 1997) and the Disability Follow-up to the Family Resources Survey (Grundy and others, 1999).

During the past 25 years, two major changes have occurred in the patterns of household and family formation (as identified in the Introduction). Marriage and childbearing are occurring increasingly later in people’s lives, and there has been a dramatic increase in childbearing outside marriage. These two changes are likely to affect family life dramatically. The way in which men and women form and dissolve their families has direct consequences on the way in which they allocate their time inside the household (see Chapter Three on the role of partners within families) and in the labour market (see Chapter Four on employment). The allocation of time within and outside families is also likely to shape the situations that men and women face when forming and dissolving their families. This chapter will argue that the two major changes in the patterns of family life can be primarily accounted for by the large increase in the tendency to cohabit in first partnerships (rather than marry immediately). In doing so, it will analyse when and in what way young people leave their parental home, when they enter their first partnership and whether it is a cohabitation or marriage, the stability of cohabiting unions, repartnering after cohabitation dissolution and the timing of motherhood. In the light of the importance of cohabiting unions in the emergence of these major changes in family formation patterns, the chapter also analyses who is likely to cohabit in their first partnership, and it investigates the factors associated with the dissolution of cohabiting unions and their conversion into marriage.

Why should discussion of the family be of interest when we look at political values or behaviour? It is unlikely that voting predispositions form a part of the ritual of courtship. The family as such cannot have an opinion. It cannot vote as a unit – for this purpose it has to break up into its individual membership. Neither politicians nor political analysts have tended to view the family as a factor in politics. For the politician thinking of politics the family is only of an interest – if with a dash of wishful thinking – as a moral force. For the political analyst, unaffected by wishful thinking, the family holds no obvious interest at all: electoral preferences are measured in terms of individual characteristics. For instance, older people, people from a relatively high social class, non-trade union members – these people are predisposed towards the Conservative Party. The family is merely in the background, providing some of the characteristics, such as tenure or total household income, that might be used to predict individual political values.

However, there are several very good reasons for taking a look at the family’s values and voting patterns:

• if real circumstances (income, jobs, and so on) influence individual opinions or political behaviour – and the extent to which this is the case can be tested empirically – then, because many of the factors that impinge on individuals vary by family circumstances (for example, a person remains in employment but their spouse or partner loses a job), the family can be seen as a possible mediator of values and voting;

• to what extent the family is a crucible for the formation or maintenance of particular political views is of great interest; conformity within the family over time is likely to mean that family members influence each other in their political and other persuasions;

• related to the above, general changes in family structure could have an impact on the transmission of political values over time, and thus the family itself might have an impact on the political system.

The panel survey provides a moving picture of people and families. It could be thought of as a form of timelapse photography in which we see beards sprouting or hairlines receding, in which partners and children move into the frame and out of it, in which the decoration and furniture become, as we view successive snaps, more lush and expensive or increasingly pinched and shabby.

The book began by thinking about social change in the aggregate over recent decades, comparing ‘cross-sectional’ survey evidence from the 1970s through to the 1990s. However, in subsequent chapters the focus on change over time tended to fall away – despite the fact that longitudinal data from the BHPS has been used throughout. The lives of the same people at successive points in time have been analysed, revealing the changes in conditions of life that they experience. However, these individual experiences of life events do not necessarily add up to social change viewed in the aggregate. On the contrary: the micro-dynamics of life, the processes through which individuals’ and households’ circumstances are maintained or transformed from year to year – what might be thought of as life chances – may simply be the best, the most informative and the most powerful way to describe the current workings of the society. Individual dynamics can sum to social statics.

The contrast between micro-dynamics and macro-statics can be illustrated by thinking about the ‘movement’ of a river. If you stand on the bank, you can see the water flowing downstream. If you come back a year later, the water passing by is a completely different set of molecules to the ones you saw last time.

During their working lives individuals can experience key labour market transitions, and it is these that are the focus of this chapter. Data from the first seven waves of the British Household Panel Survey and the lifetime employment and job histories are used to study changes in economic activity such as the transition from school to work, unemployment experiences and retirement. Career progression is also investigated by analysing the length of time people remain in the same job, career mobility, and transitions into and out of part-time and self-employment. By applying longitudinal data to these analyses, it is possible to identify those lifetime and job-related events that influence subsequent labour market changes – such as losing or gaining a job, or being promoted – and thus have direct policy relevance.

Although the labour market is complex, for the purpose of this study individuals’ working lives are categorised into three broad stages that correspond to major life-cycle events: the transition from school to work; labour market experiences over working lives; and entering retirement. These stages do not work in isolation. However, this approach provides an analytical structure to a large and wide ranging set of issues, allowing broad patterns of labour market behaviour to be established. The first stage, concerning initial labour market experiences on leaving full-time education, for many coincides with leaving the parental home and moving away from their parents’ region of residence (detailed later in Chapter Six). People leave education at different ages: some leave at 16 (at the earliest legal opportunity); others move into further education and enter the labour market at 18.