Keywords

1 Introduction

Beginning with Durkheim’s sociological research on the socialization of children (Durkheim 1922) and Parsons’ research into how young people’s acquisition of adult status is affected by spending more time in education (Parsons 1942), the issue of transition to adulthood has been widely discussed. There has been a shift from analysis based on age groups (Eisenstadt 1956; Bourdieu 1980), which has clear limitations, towards analysis that considers transition from youth to adulthood, taking into account successive stages of life history (Galland 2000; Bigot 2007; Toulemon 1994). The aim of this chapter is not to discuss the concept of acquisition of adult status or to identify which events and which sequences play a part in defining adulthood, but rather to show the changes that have taken place over generations in the timing and sequences of the first life events viewed by the literature as the markers of transition from youth to adulthood (de Singly 2000; Galland 2000; Villeneuve-Gokalp 2000; Prioux 2003; Rougerie and Courtois 1997).

These events include completion of education, entry into first employment, departure from the parental home and, finally, the first two stages of family formation—entry into first union and birth of a first child. There are numerous studies of these transitions, but they have rarely looked at all the stages from an event-history perspective.

As we shall show, the complexity of an analysis that takes into account five life-history events and their sequencing over time is probably one reason for this absence of research. However, issues of theory and methodology are important; tackling both provides better explanations of the interactions between family, occupational and educational histories. Some authors have pointed out the difficulty of analysing this transition to adulthood (Courgeau 2000; Galland 1995, 2000; Bigot 2007).

The approach most often used by demographers is to describe the prevalence and timing of a phenomenon, then to analyse explanatory factors such as behaviour or trends over time or generations. In studies of transitions or changes in status, like the first events in family formation (first marriage and first birth), demographic event-history approaches or related duration analyses have extended the scope of socio-demographic behavioural studies by introducing the interaction of phenomena over time (for example, Allison 1985; Courgeau and Lelièvre 1989; Bocquier 1996). However, analysis of the sequences and interactions of several events in the life course is still difficult to model, although it remains a favoured approach for highlighting differences in trajectories of transition to adulthood. The position and timing of these first events in the life history not only shed light on the changing conditions in which the transition to family formation takes place, but also help to identify more accurately the changes in the life course that have occurred over the generations. How have longer education and widespread female labour force participation modified the sequence of first events in the life history? How have the first two stages in the process of family formation—entry into first union and birth of a first child—been affected by these changes? Is there a convergence in the nature and timing of transition to adulthood for both men and women?

Finally, can we conclude that the trajectories of these first stages in the life history are becoming more socially diverse? Our analysis of the transitions between these events is based on data from the ERFI survey covering 10,079 men and women aged 18–79.

2 Gradual Postponement of Family Formation

2.1 Is Age at Childbearing the Sole Factor Delaying Transition to Adulthood?

The postponement of the transition to adulthood in France is a well-known phenomenon. The age at which the various steps towards family formation are taken—from completing education through to birth of a first child—has risen steadily over the generations (Bozon and Villeneuve-Gokalp 1995; Galland 2000).

These changes have taken place in a context of shifting dynamics with regard to family formation. Marriage histories are increasingly complex, and the forms taken by unions have become more diverse. Direct marriage has become rarer due to the emergence of non-marital unions, whether cohabiting or not, some of which are officialized after several years, notably through entry into a civil partnership (pacte civil de solidarité, or PACS). Similarly, union outcomes have changed, with more union dissolutions—whether through divorce or separation—occurring increasingly early after union formation (Prioux 2003).

In the different stages of transition to adulthood, the postponement of first births seems to be a constant over the generations. This delay in childbearing, to which the 1950s birth cohorts contributed significantly, has deferred the commonly recognized final stage of transition to adulthood (Daguet 2000). Our aim is not to debate whether the five first life-history events are all necessary stages on the path to adulthood (Box 1). However, some explanations will be suggested with regard to changes in the conditions in which each of the five steps are realized (Box 2). Analysis of their timing clearly highlights the general trend in male and female models of transition to adulthood.

Box 1. The Five Stages of Transition to Adulthood

The information available from the survey questionnaire enabled us to study the timing and the sequences of the five life-history first events that traditionally form the focus of the literature on transition to adulthood: completion of education, entry into first employment, departure from the parental home, entry into first union and birth of a first child. All of these are covered in the ERFI survey, although their definitions sometimes differ from those used in other surveys (Sebille and Régnier-Loilier 2007). Thus, “completing education” means the first break of at least a year in initial education (school or university) or in training for a vocational qualification. Similarly, for the concepts of departures from the parental home and of entry into first employment, the minimum duration is at least 3 months, whereas in other French surveys—the 1992 Enquêtes Jeunes (Youth Survey) and the 1997 Jeunes et Carrières (Youth and Careers Survey)—the minimum was 6 months. Although the criteria used for defining parental home-leaving, for example, differ between surveys and may lead to over- or under-estimation of timing across surveys, they do not significantly alter the sequences of events, and it is still possible to compare family, occupational and educational histories. In addition, in the ERFI survey, “parental home-leaving” means the first continuous period of at least 3 months not living with one’s parents—but irrespective of financial or residential independence from them. Under this definition, departure from the parental home can be followed by a return there for an unspecified length of time. The definition of entry to first employment is also dependent on staying in the job for a minimum of 3 consecutive months. This restriction excludes the most insecure first jobs from the analysis and focuses on long-term first jobs. Finally, the definitions of age at first union and age at birth of a first child are those traditionally used among demographers. Thus, “first union” refers to the first period living with a partner under the same roof for at least 3 consecutive months.

The biggest change is the longer interval between union formation and first birth, especially from the 1970s onwards. This has significantly delayed transition to adulthood and reduced the age differences between men and women in the transition to childbearing. Such findings prompt further questions about the convergence of timing and of explanatory models of male and female fertility. Improved birth control (notably thanks to the use of modern contraceptive methods) associated with young couples’ desire to plan their families rationally has contributed to this fertility postponement. But they are in a position to make these changes only because of the favourable context whereby people spend longer in education and attain higher levels of qualification (Desplanques 1996; Daguet 2000). As median ages at the five stages show (Fig. 2.1), age at first union has been relatively stable over the long term (between 22.5 and 24 years of age for men; between 21 and 23 years for women). Nevertheless, this pattern has been marked by changes in the social and economic climate and in the nature of entry into union (Festy 1971; Prioux 2003). Longer educational trajectories and rising unemployment—both manifest from the early 1970s onwards—have often been suggested as explanations for these trends, notably for the postponement of first marriage in post-war generations (Robert-Bobée and Mazuy 2005). It is true that deferred completion of education, which is linked to wider access to secondary and higher education, represents one of the most marked long-term changes in transition to adulthood. It has come hand in hand with a major socioeconomic transformation: the need for skilled labour and the increasing value placed on educational qualifications have reshaped the conditions for labour market entry (Estrade and Minni 1996). The 1940s birth cohorts, who experienced the end of the post-war boom period (1947–1976), were the first to significantly delay their entry into first employment; this trend also affected the mid-1950s birth cohorts, who were hit by the 1974 oil crisis.

Fig. 2.1
figure 1

Median ages by group of cohorts (in years). (Coverage: men and women in the 1926–1979 birth cohorts (8925 respondents); Interpretation: half the men in the 1926–1929 birth cohorts had completed their education by age 14. Source: INED-INSEE, ERFI-GGS1 2005)

Full size image

Box 2. Methods and Interpretation of Results

Three data analysis methods were used. The first relied on descriptive event-history analysis of the five events used as markers of transition to adulthood. In this regard, analysis of median ages taken from life tables establishes a picture of the trends in each of the events over the generations. The second method was a descriptive analysis of the sequences of the same events viewed pairwise. This method identified chronological changes in events within life courses with the aim of shedding light on the way in which people go through the process of transition to adulthood. Lastly, we used regression models (logit) to estimate which components were likely to explain the chronology of paired events. By introducing the respondent’s father’s occupational category alongside the respondent’s level of education and the prior occurrence of other events in the respondent’s life history, we can ascertain whether the pattern of these transitions differs by subpopulation, and if this pattern matches a particular sequence of steps taken that is revelatory of the conditions in which it was accomplished. These models were applied separately to men and to women so as to highlight possible differences or similarities between their trajectories. Finally, in order to reveal generational changes, the same models were applied separately to three main groups of birth cohorts: older, pre-1945 birth cohorts (specifically, 1926–1944), intermediate, post-1945 birth cohorts (1945–1959) and younger, post-1960 birth cohorts (1960–1974).

While differences between trajectories can be demonstrated, the limitations of this approach must be borne in mind, particularly when interpreting the conditions in which these transitions to adulthood occur. The fact that people may follow the same life trajectory does not mean that the conditions in which they do so are identical. For example, young people may leave their parents’ home because they have money of their own—a sign of economic independence—but also because their parents provide financial support. Similarly, any comparison between birth cohorts must take into account possible reporting or recall errors, notably among older, pre-1940 birth cohorts. Lack of precision in the dating of events, or ex-post reconstruction of life histories based on the respondents’ subjective colouring of their past may be sources of bias.

In this context of postponement of the first life-history stages, age at leaving home is an exception. Despite the deteriorating job market, age of departure from the parental home has remained relatively early by comparison with European countries overall (Régnier-Loilier 2006). In fact, in the cohorts born since the Second World War, women’s age at leaving home has fallen. Fewer women leave the parental home to marry and young women are freer from the constraints of family control; these changes have undoubtedly favoured earlier female departure from the parental household. Similarly, widespread access to higher education and a rise in family support for young people who leave home have led to the emergence of new forms of home-leaving (multiple residence or “commuting” between households, for example). These have played a major role in maintaining an early age of departure from the parental home. More and more young people live as singles for a while, notably as students, without a partner or child but supported by their parents (Arbonville and Bonvalet 2006). These conditions, associated with less secure occupational and marriage trajectories, mean that the first departure from the parental home is a less irreversible, final break for the younger generations.

2.2 Closer Timing and a Less Well-defined Order of Transition Stages

Although the general trend is towards later timing in transition to adulthood, age-specific trends in first life-history events sometimes differ for men and for women, for different sets of reasons. Thus, while women now stay in education much longer than men, this has not produced significant gender differences in the first stages of family formation. Similarly, the age gap between women and men at entry into union has remained stable, with men also forming unions later. Men’s commitment to achieving economic security is often presented as the explanation for later male timing (Courgeau 2000), and this raises questions about the predominant rationale among the younger generations nowadays. Women now have widespread access to employment and the early years of occupational trajectories are increasingly insecure: are men and women affected in the same way by changing conditions of passage through the stages of transition to adulthood? There is abundant research showing that access to a secure job comes after several years of transition between completing education, insecure employment (temporary jobs, government employment schemes) and periods of unemployment (Poulet-Coulibando and Zamora 2000; Givord 2005; Nauze-Fichet and Tomasini 2005). Has the consequence of these changes in the most recent generations been to blur the order in which the different stages of transition to adulthood occur?

For men, lack of employment associated with spending more time in education has always been recognized as a barrier to entry into union and to the birth of a first child. For women, in contrast, the relationship between occupational trajectory and family formation has only gradually become more clear. With the growing role of women as family breadwinners, female labour market participation has emerged as an element in the process of family formation (Barrère-Maurisson and Marchand 2000; Meron and Widmer 2002; Pailhé and Solaz 2006). But is it accurate to talk about a convergence between male and female models of transition to adulthood, when there are other factors at play—for example, the nature of parental home-leaving and the conditions in which it occurs?

3 Do Men’s and Women’s Trajectories Differ?

3.1 Postponement and Transition Complexity

Consistent with earlier work on the topic, postponement of transition to adulthood is confirmed for both men and women. Passage through the first events of adult life has always occurred later for men, however, and has taken place over a longer period across the generations. These changes in timing are explained by (i) delayed completion of education and entry into first employment and (ii) later age at birth of the first child (Fig. 2.1).

In that regard, our research findings complement Olivier Galland’s analysis (2000) of the cohorts born in the 1960s. In fact, the period of transition has been growing shorter since the mid-1920s birth cohorts. This observation is particularly interesting because it allows us to integrate the effects of socio-demographic changes into our analysis of trajectories followed by the French population over the last 50 years.

The cohorts born in the years 1955–1959 appear to be pivotal in this transition trend. In terms of median ages, pre-1950s male birth cohorts and pre-1955 female birth cohorts went through the first stages in their adult life history in a regular sequence. Completion of education and entry into first employment came before departure from the parental home, followed by entry into union and birth of a first child (Fig. 2.1). But this apparently ordered succession of events conceals a much more heterogeneous process, due primarily to the presence of men and women who did not experience all five events. For example, when the oldest female respondents were asked about entry into employment, 24 % in the 1926–1929 birth cohorts and 17 % in the 1930–1934 birth cohorts reported that they had never worked. The, cohorts born after the mid-1970s, for their part, are still in the course of transition. Indeed, 57 % of women and 69 % of men in the 1975–1979 birth cohorts have not yet passed through the first two stages of family formation—entry into union and birth of their first child.Footnote 1 Thus, men and women who have followed the sequence from completing education to birth of a first child by successively finding a first job, leaving the parental home and entering a union are rare. It is among the oldest (1926–1929) male birth cohorts that the proportion who experienced transition to adulthood in this order is highest (14 vs. 5 % for the 1965–1969 female birth cohorts). It should be borne in mind that the oldest birth cohorts were also those where imprecise dating of events and retrospective reconstruction of a “natural” model of transition to adulthood might have led to biased results. By contrast, with age at parental home-leaving remaining around at 20 years, and with longer educational trajectories and later entry into first employment, departure from the parental home is confirmed as the first event in the transition to adulthood experienced by the younger birth cohorts (Fig. 2.1).

3.2 More Time in Education: A Factor of Change

Detailed study of timing shows that, for men, age on completing education and age at entry into first employment have rapidly converged. Their need to gain economic independence as soon as possible may partly explain this trend, since it is a prerequisite for pursuing the other life-history stages. The gap between completing education and entry into first employment has always been small—on average, less than a year—which explains why there has been little change in this trend over the generations. In contrast, for women, the gap has changed much more significantly. In particular, the pivotal female cohorts born after the Second World War, whose age on completing education increased, saw a considerable reduction in the mean length of time separating completion of education from entry into first employment.Footnote 2 This convergence is a consequence of the increasingly frequent inversion of these two events. In fact, women and—especially—men increasingly start their first job before completing their education (34 % of men and 29 % of women in the 1965–1974 birth cohorts).Footnote 3 The trend is especially marked in the youngest birth cohorts faced with an uncertain occupational trajectory that began before they completed their higher education.

Another consequence of longer educational trajectories is that the conditions of departure from the parental home have changed, with a gradual shortening of the interval between completing education and leaving home (Fig. 2.2). In the older birth cohorts, the mean interval between completing education and leaving home was more than 4 years, while for the latest birth cohorts (1970–1974), the figure has become negative. Here again, the mid-1950s birth cohorts mark a turning-point, notably in male and female models of parental home-leaving. Before the 1950s birth cohorts, women left the parental home before completing their education less often than men did. The social control exercised over girls, who left home only when they married, and the short duration of their education explain why fewer women than men pursued their studies away from the parental home. Starting with the cohorts born after the mid-1950s, however, women began to throw off these constraints. The length of their education increased more markedly than for men, and larger numbers of women left home before the end of their school or university career. Despite these differences, male and female models of transition between completing education and departure from the parental home seem to have converged in the younger generations. In the 1970–1974 birth cohorts, nearly half the respondents had left home before completing their education. This increasing homogeneity between male and female trajectories undoubtedly originates from the widespread rise in access to higher education and the development of family support networks. In the younger birth cohorts, many went off to university towns or cities to pursue higher education, thus leaving their parents’ home—whether permanently or temporarily—before completing their education. For men, military service might have affected trends in age at departure from the parental home and at family formation. For the older birth cohorts, military service lasting 24 or even 30 months (notably for those who fought in the Algerian War) may have delayed their union formation, while for the younger birth cohorts who reached adulthood after national service had been abolished in 1996, departure from the parental home could be put off for longer.

Fig. 2.2
figure 2

Sequencing of completion of education and parental home-leaving, and mean interval between the two events. (Coverage: histograms: men and women who have completed their education and left home (8909 respondents); curves: men and women who left home and who may have continued their education (9198 respondents); Interpretation: histograms: men in the 1940–1944 birth cohorts completed their education on average 3 years before leaving the parental home; curves: 30 % of men in the 1965–1969 birth cohorts left home before completing their education. Source: INED-INSEE, ERFI-GGS1 2005)

Full size image

3.3 Leaving Home Before Entering Employment

The hypothesis of increasing parental support in achieving the first stages of adult life seems to be confirmed by analysis of the sequence of departure from the parental home and entry into first employment. The interval between entry into employment and parental home-leaving has decreased steadily over the generations, especially for men. Until the mid-1950s birth cohorts, it was not possible to leave the parental home, especially to form a family, before starting work, so men waited until they had a job before moving out. But, starting with the birth cohorts that were exposed to the economic crisis in the 1970s, the use of this strategy seems to have declined. Furthermore, men and women were both affected by the same trend. Between the 1950–1954 and the 1970–1974 birth cohorts, the proportions leaving the parental home before entering their first job rose from 26 % for both sexes to 45 % for men and 53 % for women. Parental support for leaving home, and young people’s increasing difficulty in finding a secure job may partly explain this trend.

3.4 Family Formation: A Stage that is Getting Longer

Among these changes in the early sequences of adult life, the timing of entry into first union and birth of a first child have changed over the generations—as have the conditions in which these stages occur. But, unlike completing education and entering employment, which have been postponed, age at entry into first union and age at birth of a first child have not followed the same trend. So what roles might later completion of education and later entry into first employment have played in the process of family formation?

3.4.1 A Qualified Effect of Spending Longer in Education

As in the case of parental home-leaving, longer education does not seem to have significantly affected the timing of first union. In fact, despite later age on completing education, union formation has been delayed only slightly. The mean interval between completion of education and entry into first union has even shortened significantly over the generations (Fig. 2.3). Between the late 1920s birth cohorts and those of the early 1970s, this mean interval was divided by a factor of 3.2 for men and by a factor of 6.7 for women.

Fig. 2.3
figure 3

Sequencing of completion of education and entry into first union, and mean interval between the two events. (Coverage: histograms: men and women who have completed their education and entered a first union (8132 respondents); curves: men and women who have entered a first union and who may have continued their education (9198 respondents); Interpretation: histograms: men in the 1926–1929 birth cohorts completed their education on average 11 years before entry into first union; curves: 24 % of men in the 1970–1974 birth cohorts entered a first union before completing their education). Source: INED-INSEE, ERFI-GGS1, 2005

Full size image

This phenomenon not only demonstrates that completion of education and entry into union have moved closer together, but also reveals a trend towards a sequence reversal—among women especially. Thus, in the 1970–1974 birth cohorts, over 30 % of women entered their first union before completing education, compared with under 10 % of women in the pre-1945 birth cohorts. But does this result confirm that there is no interaction between spending longer in education and entry into union? And does it confirm that men and women follow contrasting transition models? Maria Winkler-Dworak and Laurent Toulemon’s recent studies (2007) of the influence of education and employment on the process of family formation provide an explanation for these interactions and confirm the complexity of their effects. Widespread access to higher education appears to have reduced the influence of spending more time in education on the postponement of entry into union, notably for women. Nevertheless, it does not preclude an interrelationship between remaining in education and entry into union. Indeed, as these authors have shown, access to higher education reduces the interval between completion of education and union formation.

3.4.2 For Women, Entry into Union is Less Dependent on Being in Employment

In contrast, later entry into employment seems to be more closely linked to trends in timing of entry into first union, but differently for men and for women. In the male survey population, the mean interval between entry into employment and entry into union was more than 6 years, up until the mid-1950s birth cohorts who were affected by the 1970s economic crisis. It is only in the youngest birth cohorts that entry into employment and entry into union have moved closer together (Fig. 2.4).Footnote 4 In fact, the proportion of men who formed a union within 6 months of entering the labour market increased in the cohorts born after the mid-1960s. Could it be said that the need to have a job before entering a union—a prerequisite for marriage in the older generations—has been abandoned by younger generations of men? Winkler-Dworak and Toulemon (2007) give a more nuanced picture. In fact, having a job is still a determinant of entry into union. This is increasingly true for the younger generations faced with a more difficult job market and whose age at entry into first employment has increased. Being jobless is known to be a major obstacle to male union formation. Two concurrent phenomena are affecting men. The first is the established role of entry into employment in the timing of entry into union. The second is the rising trend towards a model in which men enter their first union before starting their first job, is if they are unable to delay the first stage in family formation, even though they have not yet begun their working career.Footnote 5 It is still too soon to verify this hypothesis; but the early 1970s birth cohorts are setting the trend.

Fig. 2.4
figure 4

Sequencing of entry into first employment and entry into first union, and mean interval between the two events. (Coverage: histograms: men and women who have had a job and a first union (8004 respondents); curves: men and women who have had a first union and who may not have had a job (8334 respondents). Interpretation: histograms: men in the 1935–1939 birth cohorts entered their first job on average 7 years before entering a first union; curves: 18 % of men in the 1970–1974 birth cohorts entered a first union before entering their first job). Source: INED-INSEE, ERFI-GGS1, 2005

Full size image

For women, too, entry into employment and entry into union have moved closer together; but the variation is greater than for men. The mean interval between entry into union and entry into employment averaged more than 2.5 years for the pre-1960 birth cohorts, but fell to approximately 1 year for women born after the mid-1960s (Fig. 2.4). This closer timing was, as for men, accompanied by changes in the sequences in which the two events occur. The cohorts born after the mid-1950s, the first to be exposed to the economic insecurity of the 1970s, much more frequently delayed their entry into first employment, and an increasingly large number entered a union before finding their first job.Footnote 6 This result may confirm the hypothesis that female entry into union is less dependent on being in employment. As in the case of men, not having a job may not be an obstacle to entering a union.

It is clear that there is a complex interaction between educational and occupational trajectories, on the one hand, and entry into union, on the other. For students who spend a long time in higher education, for example, entry into union may be favoured by the economic security of having a first job.

3.4.3 Birth of a First Child: A Later and More Isolated Stage

Despite the diversity of trajectories and of interactions between the first events in men’s and women’s adult life histories, we observe that a longer interval between the first two stages in the process of family formation has become a constant. As might be expected from the trend in median ages at entry into first union and at birth of a first child (Fig. 2.1), the mean interval between these two events has increased steadily over the generations, both for men and for women. The mean gap was less than 1 year for the 1926–1929 birth cohorts, but rose to more than 4 years for men and 3.5 years for women in the cohorts born from the 1960s onwards. In postponing their first birth, the 1950s and 1960s birth cohorts became the first to increase the interval between entry into first union and birth of the first child. These changes are explained in part by more modern fertility-related behaviour, linked in particular to the legalization of contraception. These birth cohorts were the first to benefit from laws that encouraged birth control and later childbearing (the 1967 Neuwirth Act, the 1975 Veil Act). On the other hand, first birth postponement seems to be linked only marginally to the emergence of complex union trajectories. Thus, among men and women in these cohorts with experience of a first union and a first child, only 5 % have entered a second union during their life course. This phenomenon developed from the 1945 birth cohort onwards.Footnote 7 However, the emergence of less stable first unions, which prove to be the first in a succession, could partly explain this dissociation between the first two stages in family formation. But there is somewhat more to this phenomenon. Whereas some 7–8 % of the post-war birth cohorts (1945–1964) reported having their first child in a second union, this applies to far fewer respondents in the younger birth cohorts.Footnote 8 The hypothesis of intermediate unions before the first birth must therefore be excluded. Delays in achieving economic security and a desire to “take time out” before the arrival of a first child would offer a more likely explanation (Régnier-Loilier 2007; Mazuy 2006).

3.4.4 Is the Timing of Male and Female Fertility Converging?

As we have shown, longer educational trajectories and later entry into employment may have affected the timing of men’s and women’s entry into union differently; however, they are often presented as the main factors in the postponement of childbearing for both men and women. The fact that the mean intervals separating completion of education and entry into first employment from the birth of a first child have remained stable for both men and women seems to confirm the hypothesis of a similar postponement pattern for both sexes (Fig. 2.5). Longer time spent in education, associated with difficulty getting a first job, may lengthen the time needed by both men and women to establish the right conditions for having a first child (Robert-Bobée and Mazuy 2005; Winkler-Dworak and Toulemon 2007).

Fig. 2.5
figure 5

Sequencing of entry into first employment and birth of a first child, and mean interval between the two events. (Coverage: histograms: men and women who have had a job and a first child (6830 respondents); curves: men and women who had a first child and who may not have had a job (7153 respondents). Interpretation: histograms: men in the 1926–1929 birth cohorts entered their first job on average 11 years before the birth of their first child; curves: 3 % of men in the 1970–1974 birth cohorts had their first child before first entry into employment). Source: INED-INSEE, ERFI-GGS1, 2005

Full size image

Although family support has certainly enabled the younger generations to leave the parental home earlier and has sometimes helped them to move in with a partner, it does not seem to provide sufficient economic security for childbearing. Thus, even though proportionally more women than men have their first child before completing their education and entering their first job, the tendency towards convergence between male and female models seems to be confirmed. For both men and women, a secure employment status seems to be a prerequisite for the second stage of family formation (Meron and Widmer 2002). Thus, far from changing, the sequence of first events in family formation with completion of education and entry into first employment is strongly established: the birth of a first child takes place at an ever later date after entry into union, completion of education and entry into employment.

4 Are Transitions Socially Marked?

Spending longer in education and postponing entry into first employment have affected the sequencing of men’s and women’s first life-history events and the conditions of transition to adulthood. More and more people in the younger generations now leave the parental home and enter a union before completing their education.

4.1 Diverse Trajectories

These sequencing changes have been accompanied by the transformation of men’s and women’s access to employment and to economic independence. Although the ERFI survey data do not allow us to assess the effects of this on the process of transition to adulthood and on family formation,Footnote 9 later entry into first employment seems to have altered the context in which the other events take place. Thus, both men and women increasingly leave the parental home before entering their first job. Thanks to the diffusion of family support models and the residential mobility associated with education, the younger birth cohorts have left home before any other event in their life history has taken place. New kinds of life course have come into being: that, for example, of students living away from the parental home who find a first job to acquire some economic independence before completing their education or forming a family. Such trajectories have arisen in the younger female birth cohorts in particular, reflecting a profound transformation in the status of women within society and the family. Their acquisition of greater residential, economic and social independence represents a major social change, from which the women in the pre-war birth cohorts were unable to benefit. Finally, the increase in age at entry into first employment—which, since the 1970s, is due mainly to young people’s difficulties in finding work and in achieving long-term economic independence—has had the knock-on effect of reducing the waiting time between entering employment and forming a family.

4.2 More Heterogeneous Transitions Today

The patterns outlined above portray the changes that have taken place in the first stages of men’s and women’s life histories, but they do little to reflect the social heterogeneity of these transitions and of the various transition trends. Completion of education and entry into first employment have been postponed in tandem; but has this led to greater homogeneity of transitions? We observed that in the cohorts born from the 1960s onwards, entry into employment, parental home-leaving and entry into union were experienced increasingly early in the life history—notably, even before completion of education. Similarly, men and—especially—women increasingly frequently entered into union before starting a first job. Unfortunately, it is difficult to know if these trends emerged for consistent reasons or if, on the contrary, they reflect transitions that are specific to different social categories.

This chapter offers a preliminary response to the question of the social diversity of transitions and transition trends. We approach it by ascertaining, for four of the transitionsFootnote 10 described here, whether social origin and level of education played a differentiating role in shaping the life courses of older generations—and whether they still do so in more recent generations. Similarly, it is interesting to analyse whether the occurrence of another life-history stage prior to the sequence studied here could be influencing transitions. This approach highlights the fact that new social disparities in the sequencing of events have emerged for both men and women over the generations.

4.2.1 Leaving Home and Starting Work Before Completing Education

Increasing numbers of men and women enter employment or leave home before completing their education. The models of Figs. 2.6 and 2.7 relating to probabilities of entry into employment and parental home-leaving before completion of education demonstrate the social heterogeneity of transitions. Even though levels of qualification have risen over the decades in all social classes, education remains discriminatory for both men and women. Thus, for both the older and younger birth cohorts, remaining longer in education leads to a commensurate increase in the probabilities of entry into employment or of parental home-leaving before completion of education.

Fig. 2.6
figure 6

Probability of entering first employment before completing education (β coefficients of the logit model). (Interpretation: for men, entering first union before entering first employment reduces the probability of starting a first job before completing education (negative coefficient). Source: INED-INSEE, ERFI-GGS1 2005)

Full size image
Fig. 2.7
figure 7

Probability of leaving the parental home before completing education (β coefficients of the logit model). (Interpretation: for women, having at least 2 years of higher education increases the probability of leaving home before completing education (positive coefficient). Source: INED-INSEE, ERFI-GGS1 2005)

Full size image

The role of social background is more complex and relates to somewhat different logical connections between the two transitions under consideration here. In the first place, the occupational category of the respondent’s father remains a determining factor of departure from the parental home before completing education (Fig. 2.7). Men and women from the families of farmers, self-employed, clerical or manual workers are less likely to leave home before completing their education than those with fathers in higher-level or intermediate occupations. For the same level of educational attainment, these results seem to verify the hypothesis that, where the father is in a higher-level or intermediate occupation, the family support network acts in favour of children leaving the parental household before completing their education. There thus appears to be a real persistence of models of transition between parental home-leaving and completion of education by social background, since leaving the parental home before completing education tends to be the preserve of better-off families whose children go into higher education.

In contrast, for the sequence of events whereby young people enter first employment before completing their education, the role of social origin over the generations is an obvious one. For men and women in the older birth cohorts (1926–1944), social origin does not seem to have been significant. For men, this lack of significance persisted until the late 1950s birth cohorts. For women, on the other hand, being the daughter of a farmer or a clerical worker in the 1944–1959 birth cohorts was an obstacle to entering employment before completing education; for men, this obstacle began to emerge for the sons of self-employed and manual workers in the 1960–1974 birth cohorts. Unlike the first transition we studied, social background appears to be discriminatory in the trajectory leading from employment to completion of education over all cohorts. It seems to correspond strongly to periods in the economic cycle and to obstacles in finding employment which faced women and then, later, men.

Finally, these transitions to first employment or parental home-leaving before completion of education very obviously relate to particular life history dynamics. Thus, early entry into union before entry into employment has always been an obstacle to entering employment before completing education. For men and women in both the older and the younger birth cohorts, forming a first union before entering the labour market strongly reduces the probability of entering employment before completing education. This demonstrates that early union formation is most likely to form part of a more “traditional” trajectory, where completion of education precedes entry into first employment. Similarly, entering employment before leaving the parental home, whether or not in anticipation of residential independence, favours parental home-leaving before completion of education. Having one’s own income most certainly encourages earlier departure before completion of education.

4.2.2 Entry into Union Before Completion of Education and Entry into First Employment

As our analysis of the patterns of sequences has shown, people increasingly enter their first union before completing education and finding a first job, especially from the 1960s birth cohorts onwards. Analysing disparities over the cohorts reveals social differentiation in the transitions between entry into union, on the one hand, and completion of education and entry into employment, on the other.

Just as in the case of the first transitions examined, level of education is a logical factor of heterogeneity in transitions between entry into union and completion of education. Across the generations, for both men and women, a higher level of education has always favoured entry into first union before completion of education (Fig. 2.8). But the changes have arisen from disparities linked to social origin. Although this life-history sequence was rare in the older birth cohorts, it was not then the preserve of a specific social category. Even though spending longer in education has spread to all social classes, significant differences appear from the post-war birth cohorts (1945–1959) onwards. The less highly educated sons and daughters of farmers and clerical workers enter their first union before completing their education less frequently than their peers. However, this social dynamic disappears in the youngest male birth cohorts, although it continued among farmers’ and manual workers’ daughters in the post-1960 birth cohorts—evidence of a wider heterogeneity of transitions for women nowadays.

Fig. 2.8
figure 8

Probability of entry into first union before completing education (β coefficients of the logit model). (Interpretation: for the 1960–1974 birth cohorts, being the daughter of a manual worker reduces the probability of entering a union before completing education (negative coefficient). Source: INED-INSEE, ERFI-GGS1 2005)

Full size image

There is one constant in the life courses of all generations, however. Entry into employment before first entry into union reduces the probability of experiencing a sequence where entry into union precedes completion of education. This relationship between entry into employment, entry into union and completion of education may reflect two different types of trajectory. The first, which was very frequent in the older birth cohorts, corresponds to a sequence where education is completed before entry into first employment and subsequent entry into union. This type of trajectory was very frequent among men born before the mid-1950s in particular. The second trajectory is represented by early entry into employment, before completion of education and entry into first union. This trajectory has emerged especially in the younger birth cohorts, who remain longer in education and therefore enter employment before they enter a first union, thus delaying family formation.

In fact, it is clear that the relationship between completion of education, entry into first employment and entry into union lies at the heart of heterogeneity in men’s and women’s life courses. For example, in the cohorts born before the mid-1940s, educated men entered their first union before entering employment more frequently than other men did, whereas the numbers of low-educated women who went through that particular transition remained marginal (Fig. 2.9). However, there was no observable social difference in this transition. It was from the post-war birth cohorts onwards, in particular for women, that differences in trajectories appeared. A distinguishing feature for the daughters of manual workers and, later, the daughters of clerical and self-employed workers, was that only a marginal number of them entered a union before starting their first job. For men, this phenomenon was confirmed only from the 1960s birth cohorts onwards. The sons of men in higher-level occupations differed from others in that they were more likely to form a union before they entered their first job. These were undoubtedly young men in a situation where family support and parental assistance enabled them to start forming their family before completing their education and starting a career.

Fig. 2.9
figure 9

Probability of entry into first union before entry into first employment (β coefficients of the logit model). (Interpretation: for the 1960–1974 birth cohorts, being the son of a man in a higher-level occupation increases the probability of entering a union before entering employment (positive coefficient). Source: INED-INSEE, ERFI-GGS1 2005)

Full size image

These initial results by level of education and father’s occupational category show that educational trajectories have remained factors of heterogeneity over the generations. They also provide evidence of the growing role of social origin in the ordering of transitions between entry into employment, completion of education and the first stage of family formation. However, the changes affecting men’s and women’s occupational trajectories over the generations should be introduced into the analysis of these social disparities. The way in which people enter their first job has changed a great deal, and people in the younger generations face a period of instability and insecurity. These major changes have undoubtedly modified the conditions in which transitions between the first stages of life history and family formation take place.

5 Conclusion

Known trends in the timing of the first stages of adult life provide a picture of changes in the path towards entering first employment, leaving the parental home and forming a union and a family. Completion of education and entry into first employment undoubtedly remain two of the key events in this transition to adulthood. A significant rise in levels of education, especially for women, has resulted in changes to the length of time that separates completion of education from the events of family formation. Although this is not the direct, immediate cause of later entry into union, it has not encouraged young people to move quickly towards residential and economic independence. Longer time spent in education and difficulties entering the labour market have created difficult conditions for parental home-leaving and union formation. It is striking that age at leaving the parental home has not changed significantly over the generations due to a radical transformation of the conditions in which this transition occurs. Many young people receive support from their parents to move into an independent dwelling before completing their education or finding a steady job. This has in some sense compensated for a general delay in achieving economic independence among the younger generations. The economic crisis of the 1970s that followed the post-war economic boom, along with increasingly tight labour market conditions from the mid-1980s onwards, certainly made things much harder for young people trying to enter the labour market.

These socioeconomic changes, associated with the social transformations that have affected women in particular, have led to shifts in the timing of the first stages of adult life. One of the chief results has been that these first stages—completion of education, departure from the parental home, entry into employment and entry into union—have moved closer together. Only the timing of the first birth, which has always taken place later, has remained largely untouched. While it does not attempt to explain trends in the sequencing of the first stages of life history, this study highlights the emergence of new trajectories and offers a nuanced picture of the convergence of male and female life courses. It also shows that, over the generations, social disparities in transitions are becoming stronger—a reflection of complex and heterogeneous trajectories.