Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

5.1 Introduction

The transition to adulthood for both men and women has undergone considerable changes in European countries since the mid-1960s. Participation rates in higher education have steadily increased. Women have intensified their labour force participation. Traditional family related institutions have less of an impact on early life courses. Marriage and family formation are delayed. Non-marital and childless forms of living arrangements have stepped in to fill the gap. In general, highly institutionalised traditional pathways to social and economic independence have lost relevance for young men and women. They have changed their behaviour to profit from the opportunities the welfare state institutions and the economic market have to offer. At the same time, they have to follow the rules of these institutions to be successful. The need for individual autonomy and flexibility has increased.

In the social sciences, it seems to be taken for granted that an all-embracing de-standardisation of patterns of transition to adulthood, and of individual life courses in general, have been the consequences of these developments. It means that individual life courses become more diverse, that the variance of the age distribution at particular events increases, and that the heterogeneity of life states among members of certain age groups amplifies. Indeed, this view is widely supported by empirical evidence. Findings of studies in which this process has been investigated show a profound and constant trend of a dilution of formerly highly structured passages of the life course in nearly all European countries (Billari and Wilson 2001; Corijn and Klijzing 2001; Elzinga and Liefbroer 2007). Between different countries, considerable diversity can, however, be observed with regard to both the onset and the extent of this change (Billari 2004).

In the following, I will argue that this comprehensive view of de-standardisation falls short of the logic of processes of social change. There are good theoretical and empirical reasons to expect that patterns of transition to adulthood just change and new standardised patterns of this status passage in the life course arise. Reverse trends to re-standardisation can emerge. In many cases, de-standardisation can just be a transient phenomenon, occurring during periods of pattern change and resulting in new, but again highly structured patterns of life course trajectories. Therefore, the general impression of an increasing diversity of individual life courses and transition to adulthood might be “biased”.

In this chapter, I discuss this issue from a conceptual, theoretical, and empirical perspective. First, I attempt to clarify what (de-) or (re-)standardisation means. Second, I outline some theoretical approaches which support the de-standardisation hypothesis, and then argue for an alternative view from a theoretical perspective. Third, I present some empirical evidence. I present own findings for the German case and results from comparative research as found in the literature. Different trends can be identified, including cases of de-standardisation, of changing patterns without de-standardisation emerging, and of re-standardisation. Finally, I draw some conclusions. Having looked closely at what actually happens in European countries I propose that alternatives to the de-standardisation hypothesis must be considered. In particular, re-standardisation as a systematic phenomenon in demographic trends has to be considered more seriously than has so far been the case.

5.2 De-Standardisation and Re-Standardisation in the Life Course: Some Conceptual Clarifications

Before I discuss the de-standardisation issue from a theoretical and an empirical point of view I start with some conceptual clarifications.

5.2.1 The Concept of Standardisation and De-Standardisation

When referring to de-standardisation of all or part of the life course, Brückner and Mayer (2005) mean that “life states, events, and their sequence can become experiences which either characterize an increasingly smaller part of a population or occur at more dispersed ages and with more dispersed durations” (Brückner and Mayer 2005, 32f). In Table 5.1, I seek to clarify in more detail what factors the level of standardisation of all or part of the life course, such as the family life trajectory or the status passage of transition to adulthood, depends upon. Several dimensions are distinguished (cf. Huinink and Konietzka 2003). I will not comment on them in detail.

Table 5.1 Dimensions of the level of standardisation of the life course
Full size table

In this context, de-standardisation means that the level of standardisation is declining in one of these seven dimensions. This can be, for example, increasing entropyFootnote 1 of age distributions, decreasing correlations between the occurrence of biographical states or events, or increasing entropy of the distribution of types of biographical trajectories.

The definition by Brückner and Mayer primarily refers to dimensions one to five, and partly to dimension six in Table 5.1. In the literature, other measures of the level of standardisation are proposed which mainly deal with trajectories or parts thereof (dimensions six and seven). Elzinga and Liefbroer (2007) analyse the turbulence and dissimilarity of trajectories, or the variation of pathways to adulthood. Elzinga and Liefbroer (2007, 232) “consider series [meaning trajectories, JH] that have many distinct states and many state changes as being more turbulent than series with fewer distinct states and/or fewer state changes”. This concept is a measure of reversibility of life events and stability of biographical states in trajectories (dimension six in Table 5.1). The dissimilarity of trajectories and the variation of types of trajectories, which is based on a typology of typical patterns, are measures of diversity of trajectories (dimension seven in Table 5.1).

5.2.2 “Modes” of De-Standardisation and Re-Standardisation

Conceptually, one can distinguish two modes of de-standardisation:

  1. 1.

    The transitory case

    De-standardisation can be just a transitory phenomenon caused by a special characteristic of the dynamics of pattern change (pattern-transition hypothesis). It is a passing phenomenon which occurs when the shift from one pattern to another does not encompass the whole population simultaneously but part by part like in the case of pattern diffusion. During the period of change an increasing part of the population experiences the new pattern while the other part still follows the old one. Consequently during the initial phase of pattern change the diversity increases (de-standardisation) but after a certain period of time it decreases again (re-standardisation) because finally the new pattern has been established in the whole population. If the pattern shifts gradually throughout the whole population simultaneously, change without any de-standardisation can take place. This, however, is an exception.

    De-standardisation of this transitory kind might also be caused by a transitory deviation from a pattern, disappearing after a certain period of time when in the end the old pattern survives. This can happen during and after historical events which have a major age specific impact on individual life courses, such as wars or economic crises.

  2. 2.

    The irreversible case

    De-standardisation might evolve as an ongoing trend of pattern dissolution over time (de-standardisation hypothesis). In this case, de-standardisation is not reversible in the long run. It means that a certain pattern is being diluted, without leading to a new one or a return to the previous one.

    The mode of transient de-standardisation makes clear that re-standardisation is a possibility. For example, one dominant age at a biographical event is, after a time, simply “replaced” by a different one; or one dominant pathway to social independence from parents is “replaced” by another pathway. Re-standardisation means that one or more of the dimensions of standardisation in Table 5.1 is once again increasing: e.g., age patterns narrow, the connection between life events strengthens, a certain kind of new living arrangement in a certain age period is gaining dominance, or a sequence of events is appearing again. Corresponding to the definition of de-standardisation by Brückner und Mayer, re-standardisation means that life states, events, and their sequence become experiences which (again) either characterise an increasingly larger part of a population, or occur in more narrow age spans and with more uniform durations.

5.3 Theoretical Considerations

This section provides a brief overview of some theoretical arguments in the literature supporting the de-standardisation hypothesis. Some of these arguments have a particular focus on transition to adulthood. In the social sciences, the transition to adulthood is traditionally conceived of as a process in individuals’ life courses which is connected with typical transition markers, such as starting and completing education, starting a job or career, leaving the parental home, forming a union, and a family. It is directed towards the goal of gaining individual economic and social independence by the young adult.

5.3.1 Sociological and Demographic Considerations

The concept of the Second Demographic Transition (SDT) promotes the idea of a new period of demographic change affecting in particular the status passage of transition to adulthood. While the First Demographic Transition led to a highly standardised pattern of bourgeois family dynamics, labelled the “golden age of marriage”, these patterns dissolve during the Second Demographic Transition. The main trends have already been mentioned: declining family size, increasing age at family formation, increasing relevance of non-marital living arrangements and extra-marital childbearing, and declining stability of couples and rising divorce rates. These trends, it has been argued, are the result of ongoing secularisation, shifts in value orientations, and an increasing relevance of higher level goals, like self-realisation and self-fulfilment in the lives of individuals (van de Kaa 1987; Lesthaeghe and Moors 2000; Lesthaeghe and Surkyn 2007). The question of de-standardisation has not been addressed explicitly in this context. In a recent article, Lesthaeghe and Surkyn made the following prediction regarding the SDT: “The normative and institutional props of traditional union formation and household structures will systematically weaken in all societies that move in the direction of egalitarian and democratic systems governed by the respect for individual choice. This implies that other forms of union formation will expand in the wake of such ideational developments” (Lesthaeghe and Surkyn 2007, 113). No explicit arguments have been made so far that the SDT could pave the way for new patterns to emerge, such as new standards of family formation and living arrangements.

That is why many scholars draw the conclusion that the SDT should, by and large, lead to more diversity. Corijn and Klijzing (2001) use assumptions about the SDT approach and the modernisation paradigm when assessing change in the transition to adulthood. In regard to age graduation, they generally propose that the age-relatedness of markers of transition to adulthood is weakening. However, they argue that age-relatedness of career related transitions remains stronger than age-relatedness of family related transitions. This is true because education and income capacity are the most important requirements for a successful life, and the institutional regulations and the normative expectations in regard to the time span for achieving these goals are quite restrictive. They also assume that de-standardisation should not surpass the age of thirty. The level of de-standardisation between different European countries should remain different. In regard to the interrelatedness of different dimensions of the transition to adulthood, they conventionally argue for an incompatibility between education and parenthood—more so for women than for men, and weakening with increasing age, by different degrees in European countries.

Billari and Wilson (2001) also deal with the question of diversity vs. similarity in the shift towards greater de-standardisation in the transition to adulthood in different European countries. They also do not support the hypothesis of convergence between European countries. By and large, they seem to take for granted that there is a general trend towards a de-standardisation of this status passage. In any case, other alternatives are not addressed when they conclude with the following: “Rather than seeing life course changes as being driven by strongly determined trends towards country-level convergence, of either the simple form or the individualization variety, perhaps we should see recent trends as being ‘enabling’? With increasing wealth in European countries many of the material constraints on life course transitions have been reduced or removed. Thus, individuals in different societies are freer than ever before to give free rein to their preferences” (Billari and Wilson 2001, 14).

In sociology, a prominent position arguing for the change in individual life courses is based on the hypotheses of de-institutionalisation and ongoing social differentiation in late modern societies (Brückner and Mayer 2005). De-institutionalisation means “that states, stages, events, and transitions, which at earlier times were clearly differentiated [by institutional regulations; JH], are now being reintegrated or fused” (Brückner and Mayer 2005, 32). The change evolves after a period of institutionalisation and standardisation during modernisation (Kohli 2003). Currently, modern institutions like the bourgeois family system are eroding. Life courses are less institutionally predetermined. Young adults do not follow traditional social norms in planning their lives. Instead, they tend to optimize strategies in a far more flexible and autonomous way to make the most of a society that is now rich in opportunities and risks.

Social differentiation “refers to the process where the number of distinct states or stages across the life course increases” (Brückner and Mayer 2005, 33). It could be added that social differentiation also leads to a de-coupling of different dimensions of biographical states. The areas of activity and individual engagement they refer to increasingly follow their own procedural logic, with their own demands and options for individual actors.

Brückner and Mayer (2005, 35) note that processes of institutionalisation and de-institutionalisation could go hand in hand. This is likely to be particularly true in a highly functional differentiated society. While no general trend can be observed, they assume a movement towards de-standardisation and increasing flexibility in all fields of life .

Elzinga and Liefbroer (2007) start with the same concepts of differentiation and de-standardisation employed by Brückner and Mayer. They also assume that the mobility of young adults between different states has increased among recent cohorts. This hypothesis is tested by analysing family life trajectories in various European countries. The authors also propose that trajectories are less similar between young adults, and that no dominant type of family trajectory now exists. They also assume a considerable variation between different countries in Europe, anticipating that variations will be greatest in social democratic welfare states and smallest in the Mediterranean and East European countries.

Finally, an influential hypothesis in sociology proposes a new shift in individualisation in late-modern societies (Beck 1986; Shanahan 2000). Generally, individualisation means that individual actors are, in planning and mastering their life course, no longer tied to normative institutions, community related commitments, and traditional support systems. This implies a transformation from community type organised social contexts to contexts centred on the market and on formal regulation-based systems of co-ordination and integration. The “new individualisation” thesis assumes that women are liberated from the traditional roles of mother and housewife, and gain better access to the labour market. Furthermore, the modern institutions of the life course, like the bourgeois family pattern which resulted from individualisation processes during modernisation itself (Kohli 1985) lose their impact. Individual agency, flexibility, and choice gain relevance. The transition to adulthood can be expected to become more variable with the transition markers becoming decompressed (Shanahan 2000, 671), and the sequencing of markers becoming more diverse.

5.3.2 A Psychological View

An approach coming from developmental psychology also argues in favour of more de-standardisation. Arnett introduces the term “emerging adulthood” (Arnett 2000, 469), a period defined as being “neither adolescence nor young adulthood”. Sociologists would say there is some anomic situation, without specific social roles and normative expectations to be obeyed. Emerging adults are becoming independent of their parents, and social control by parents has decreased considerably, yet these young people still have not taken on the “responsibilities that are normative in adulthood”. The educational system has expanded, and marriage and childbirth have been delayed. Perhaps because they follow new age norms, economic pressure on young adults has been moderate. The result is “a high degree of demographic diversity and instability, reflecting the emphasis on change and exploration (in areas of love, work, and worldviews; JH)” (Arnett 2000, 471).

Arnett’s arguments could well support the hypothesis that there is a de-standardisation of transition to adulthood, at least as far as the level of turbulence in trajectories during this period of the life course is concerned. Accepting that de-standardisation is occurring does not contradict the notion that, in terms of age and biographical states, re-standardisation and a rather standardised way into and out of this period could also be taking place. Emerging adulthood has itself become a period of the life course where the weak norms are normatively supported. The level of diversity is therefore a function of the supply of attractive opportunities to extend exploration; here we can draw links to the issues of differentiation and societal and technological development.

5.3.3 Conclusions

All approaches, as diverse as they are, generally appear to favour assumptions of an ongoing de-standardisation of the transition to adulthood. In addition to postponement of events, their age-relatedness should have declined, biographical disorder should have increased, and it should be possible to observe a decoupling of life events during the early adult life course (Furstenberg et al. 2005). Alternatives are not discussed. From my point of view, these concepts and theoretical approaches suffer from three deficits:

  1. 1.

    For the most part, they take for granted that de-institutionalisation and differentiation lead to the dissolution of life course regimes, without considering any concrete ideas of restructuring trends.

  2. 2.

    They do not differentiate between the various modes of de-standardisation, as I proposed.

  3. 3.

    They do not consider explicitly trends of re-standardisation, even though a hypothesis pertaining to this could easily be implemented in these theories.

In discussing the de-standardisation hypothesis, a general argument could be made that social change can be conceived as a sequence of periods of stability or standardisation, and periods of more or less rapid change accompanied by (transient) de-standardisation. In addition, it could be argued that scientific, economic, and social developments lead to an increase in opportunities and abilities, which support an underlying trend of increasing diversity in human behaviour and individual life courses. Both aspects must be considered. Given the general nature of this discussion, I will not address these issues in great detail, but will instead restrict myself to some simple arguments.

De-standardisation can be conceived of as a possible structural or demographic consequence of de-institutionalisation and differentiation (Huinink and Konietzka 2003). But the question of de-standardisation is more complicated than is frequently assumed (Huinink and Wagner 1998; Brüderl 2004). De-institutionalisation may be a necessary condition for de-standardisation, but it is not sufficient. If we assume that people act (quite) rationally in a given set of living conditions, we can propose the following: Even without social norms governing social action, it should be possible to find common behavioural patterns reflecting the optimal way to deal with circumstances of action and life course planning. Therefore, de-standardisation can mean that the circumstances in which young men and women plan their lives have become increasingly diverse. At the same time, however, de-institutionalisation itself might just be a transient phenomenon occurring in a phase of changing transition patterns. De-standardisation could diminish as soon as new institutional patterns take over after a period of re-institutionalisation (pattern-transition or re-standardisation hypothesis).

In exploring the first point, it is interesting to look at the theoretical concept Blossfeld and his research group proposed in studying the consequences of globalisation on individual life courses. They argue that the restructuring of local economies and demographic change in many countries put welfare state institutions and labour markets under pressure. On the individual level, biographical uncertainties arise affecting individual life plans and instabilities. New risks in educational and occupational careers emerge as a new threat to a promising transition to adulthood (Blossfeld et al. 2005). Following an action theoretic model of bounded rationality, they state that young men and women cope with this uncertainty under the given institutional conditions. One central hypothesis is, then, “that the uncertainty generated by globalisation at the socio-structural level reduces or delays the propensity of youth to enter long-term binding commitments such as partnership and parenthood” (Mills and Blossfeld 2005, 16). However, they do not propose an increasing de-standardisation of early life courses in general, but rather suggest that there is considerable diversity caused by different institutional conditions in the mastering of increasing uncertainty.

In conclusion, we can state that, in some life domains or dimensions of transition to adulthood, irreversible trends of de-standardisation may emerge, particularly in the case of a stable interplay between de-institutionalisation and social differentiation. Sometimes de-standardisation does not occur. In other cases, de-standardisation is transient, and comes to an end when new patterns are established that are grounded on new institutions or new structural typicality in the lives of young individuals. The specific circumstances surrounding the achievement of the different goals on the way to adulthood determine what can theoretically be expected. Which version might be most probable for which dimension of transition to adulthood? This question is difficult to answer, and I cannot present an elaborated theory at this point. Instead, let us look at empirical examples that may supply us with evidence supporting one of these cases.

5.4 Some Empirical Evidence

My empirical analyses focus on four of the dimensions previously mentioned in Table 5.1. The age structure of biographical events, the interrelation between biographical events, the sequence of biographical events, and the diversity of whole trajectories. I continue to focus on the status passage of transition to adulthood.

5.4.1 The Age Structure of Biographical Events

Let us start with a comparison of the age at biographical transitions leading to social independence from parents between different cohorts. Beginning with the West German case, I look at first household formation, first consensual union, and family formation.

In Table 5.2, parameters of the age distribution at first household formation for men and women of seven West German cohorts are displayed. Men’s median age at first household formation decreased until the cohort 1959–1961, from age 26 to 23. In later cohorts, median ages rose again, though only by less than 1 year. For women, we observe a decline of the median age from 24 to 20.5 years between cohorts 1929–1931 and 1954–1956. Again, in the younger cohorts, the median age increased by about 1 year. Over the same period, the “time corridor” of leaving home narrowed over the cohorts. After a considerable decline, a slight increase in the age difference between the first and the third quartile can be observed for men in the 1971 cohort. Still, the age difference was only about 5.5 years for the youngest cohort in our sample, while it was more than 7 years in the oldest cohort. For women, the decrease in the age difference between the first and the third quartiles was even greater, falling from about 8 years in the cohort 1929–1931, to as low as around 3.5 years in the cohort 1954–1956, and then rising again to 4.5 years in the 1971 cohort. These figures suggest that there is little indication for de-standardised patterns of first household formation in terms of cohort-specific age variations in West Germany.

Table 5.2 Age at first household formation by cohort and gender in West Germany
Full size table

The same holds true if we look at the age at starting the co-residential union. In Table 5.3, results from Kaplan–Meier estimates of this age are displayed. I used data from the German Family Survey 2000. (Age is calculated as the difference between the year of the beginning of the partnership and the birth year). As can be seen in Table 5.3, there is little change in regard to the age structure of this event for men and women. No trend of age dispersion is apparent.

Table 5.3 Age at first consensual union by cohort and gender in West Germany
Full size table

Finally, let us turn to the transition to first marriage and to the first child. I focus only on West German men and women. Again, I used data from the German Family Survey 2000. The respective figures are displayed in Table 5.4.

Table 5.4 Age at first marriage and first birth of West German men and women, by cohort
Full size table

Both age at first marriage and age at first birth increased considerably over cohorts. Looking at the first marriage of West German women, we see that the age range widens over cohorts when the age at first marriage rises. The interquartile differences for women increase from 5.5 years in cohort 1944–1949 to 9.3 years in cohort 1962–1967. For men, we cannot compare interquartile differences between cohorts with the data used.

A different picture is apparent for age at first birth. Women of the cohort 1944–1949 show quite a standardised pattern of family formation at an early age. By contrast, age at first birth is quite dispersed among women of the cohort 1950–1955. It is a typical transitional cohort at a time when the trend towards delay of family formation was underway. In the younger cohort of 1962–1967, a new age graduation appears to have begun, though the delaying process was not completed. Late family formation in the late twenties was becoming increasingly normal. The interquartile difference, which can be used as measure of the level of standardisation, changes over cohorts. For the cohort 1944–1949, it is 10.1 years, while for the cohort 1950–1955, the estimate is 11.6 years. For the cohort 1956–1961, the estimate is 9.0 years, and for the cohort 1962–1967, it is even smaller (8.6 years). This shows the transitory character of the emergence of an age-dispersed pattern of family formation in the cohort 1950–1955. Interestingly, we see that, for women, the findings support the idea that de-standardisation is just a transient phenomenon caused by the age shift for family formation, but not for marriage.

Billari and Wilson (2001) did similar calculations for different European countries, allowing us to make international comparisons. They estimated interquartile differences in ages at leaving home, first union, first marriage, and first birth. I only show part of one of their tables dealing with first marriage and first birth. We see in Table 5.5 that the patterns for West Germany match results from some other countries. In nearly all countries, the age at marriage is constantly de-standardising, particularly in Sweden. In the case of age at first birth, we often observe a nonlinear trend, as could be expected given the pattern change assumption.

Table 5.5 Interquartile differences in ages at events in the transition to adulthood (years)
Full size table

5.4.2 Interrelation Between Biographical Events

Now I turn to some examples of the interrelation between the markers of transition to adulthood and the sequencing of events. Currently in West Germany, relatively few young men or women first move in together at or after marriage [Huinink and Konietzka (2006), the changing impact of union formation on leaving home in Germany. A cohort analysis of interdependent life events in the transition to adulthood “unpublished”; Kley and Huinink (2006)]. As Fig. 5.1 shows, cohabitation has nearly completely taken over the role of marriage in this regard, more so for women than for men. But moving in with a partner still is the most frequent reason for leaving the parental home in West Germany.

Fig. 5.1
figure 1

First household formation and other markers of transition to adulthood. Source German Life History Study [Huinink and Konietzka (2006), the changing impact of union formation on leaving home in Germany. A cohort analysis of interdependent life events in the transition to adulthood “unpublished”]

Full size image

It is possible to add figures on the dependency of different transition rates on leaving the parental household which nicely support these descriptive findings [Huinink and Konietzka (2006), the changing impact of union formation on leaving home in Germany. A cohort analysis of interdependent life events in the transition to adulthood “unpublished”]. I interpret this trend to be a kind of re-standardisation and re-institutionalisation following the de-institutionalisation of marriage. As has been shown in many analyses, a decoupling of first marriage and first birth can be observed in many European countries, as is the case for Germany. In Southern European countries like Italy, marriage still seems to be a prerequisite for leaving home. That might be one major reason for the unprecedented rise in the median age at leaving home there. Paradoxically, the stability of institutions is causing a change in patterns of transition to adulthood, leading to long delays in family formation.

Our comparative outlook now deals with the sequencing of leaving home, of first union (cohabitation), and of first marriage. Again, we display the results of Billari and Wilson (2001) in Table 5.6. In countries which are assumed to be prototypes of de-standardisation, we find highly standardised patterns. The estimations of Billari and Wilson show that Sweden is ahead in a re-standardisation in this regard. Other countries are on the way. In Sweden, the traditional patterns have diminished, resulting in new regularities. Formerly unconventional sequences of events now are obligatory. This is in accordance with the pattern–transition hypothesis. We do not know, however, whether this is caused by new social norms or by other factors.

Table 5.6 Sequencing between events during transition to adulthood (years)
Full size table

5.4.3 Diversity of Biographical (Sub-)Trajectories

5.4.3.1 Living Arrangements and Family Career

Relationships are arranged differently today than they were in the past. A pullback from marriage can be observed, and the age for starting a consensual union with a common household is rising, too. Living apart together and cohabitation are gaining relevance, as is living alone for a period of time. However, this does not mean that “anything goes”. Cohabitation has evolved into a highly institutionalised step or long-term alternative to marriage.

But from the longitudinal view, a moderate shift in the diversity of the trajectory of living arrangements is shown for West Germany (Brüderl and Klein 2003; Brüderl 2004). Data from the German Family Survey 2000 show that the entropy of the distribution of living arrangements among young adults after age 22 increased over time, more so in large cities than in rural areas. One reason for this trend is the increasing proportion of men and women living without a partner at this age segment. This is a transient state in most cases, but obviously it occurs by a higher rate. Brüderl and Klein also show an increasing divergence from the traditional pattern in the pathways to adulthood. This can, but by no means must be seen as evidence of de-standardisation, because new patterns of standardisation could have emerged.

To provide an international comparison, I briefly come back to Elzinga and Liefbroer (2007), who studied family trajectories of young adults in different European countries using data of the Family and Fertility Survey. They compared cohorts 1945–1949, 1950–1954, 1955–1959, and 1960–1964. As mentioned previously, they developed new measures of turbulence, the similarity of trajectories and their variation over seven ‘ideal types’ of pathways.

They found that turbulence was rather stable over cohorts in most countries, but it rose in some places, and it even declined in two countries (Poland and Italy). The similarity between trajectories declined in nearly all countries, though only slightly in some of them. The level of variation between types of trajectories was also found to be rising in most countries. In Sweden, it has been declining over cohorts. Elzinga and Liefbroer argue that, in this case, a new standard might be emerging, and they refer to re-standardisation. They conclude: “These results offer strong support to the idea that the family life trajectories of young adult women all across the Western world are becoming more de-standardized. […] the decrease in variation in family life trajectories observed in Sweden and, to a lesser extent, also in the Netherlands, suggests that new standards may be emerging in these countries. In that sense, it is not impossible that a process of re-standardization will occur in countries that have experienced the Second Demographic Transition” (Elzinga and Liefbroer 2007, 246–247).

Do these results contradict the findings of Billari and Wilson? I do not think so, assuming that turbulence is not equated with de-standardisation. Taken together, the results of these two studies confirm the assumption that the stability of trajectories during the transition to adulthood has been declining. Yet, we nevertheless observe standardised parts of sequences and transitions, mainly based on new standards of ageing and sequencing.

5.4.3.2 Apprenticeship and Labour Market Entry

Finally, we take a brief look at the transition to economic autonomy in Germany: starting and completing post-school training or apprenticeship, and starting a job or career. (In Germany, vocational education of this kind is very important: it combines apprenticeship in a company with part-time education in special vocational schools that provide the theoretical education to the company-based practical training). The de-standardisation hypothesis proposed by Arnett could prove useful in explaining these trajectories, because his hypothesis is more strongly related to processes of differentiation, and is motivated by his concept of emerging adulthood.

A certain trend towards de-standardisation can, in fact, be shown in regard to vocational training and apprenticeships. The findings of Konietzka (1999) for the West German case, which are based on data from the German Life History Study, show that the percentage of young men and women who started more than one apprenticeship increased significantly from the cohorts of 1929–1931 to the cohorts of 1959–1961. However, the percentage of men who completed more than one apprenticeship was stable over cohorts. It rose only slightly for women. This could mean that, in younger cohorts, a larger share of young men and women try out different kinds of apprenticeship before completing one that best fits their interests (cf. Arnett 2000). The increase can be a function of improving opportunities as well as higher uncertainties about their life plans. While the former would support a trend to irreversible de-standardisation, this is not true for the latter.

Konietzka is also able to show that the transition to the first job, even when it occurs in times of economic crisis, is still quite regular among West German cohorts, and is only gradually becoming fuzzy. No de-standardisation is apparent. But West Germany might be a special case when compared with other European countries, for which I cannot present similar studies.

5.5 Conclusions

The brief overview of studies of the changing transition to social and economic autonomy of adolescents and young adults shed light on both the processes of transient de-standardisation as a consequence of pattern change, and the trends towards more diversity, flexibility, and variety of trajectories during the early life course. It is true that the latter is dominant in Europe. In line with Arnett’s theory, there are hints of a more pronounced period of experimentation. We also know that phases of living alone gain relevance. The level of the institutionalisation of intimate relationships has changed, with the strength of commitment in the early stages of relationships being lower than in the past. The diversity, at least in the longitudinal perspective (trajectories), has become larger, and yet more differentiation of living arrangements can be assumed if the numerous versions of arrangements where partners are regionally mobile are taken into consideration. It should also be noted that the prevalence of highly committed living arrangements has also declined, representing yet another indication of de-standardisation.

However, this does not mean that there are no signs of new institutionalisation and re-standardisation. For example, cohabitation has virtually become the standard step into a co-residential household in a growing number of countries. Even family formation out of wedlock is increasingly becoming a standard behaviour in some countries like Sweden and East Germany. The age at which certain markers of transition to adulthood take place has changed, but in many cases not dispersed. The age structure of events like marriage and family formation has changed, but also re-stabilised.

De-standardisation might have its limits. For example, partnership, however it is realised, is still an important state to be reached quite early in adult life. New schedules for young adults might emerge. Therefore, enduring de-standardisation is not the only alternative for the future. For instance, that social norms regarding leaving home still matter, but only in a very restricted way, is nicely shown by Billari and Liefbroer (2007).

This is why empirical alternatives to the de-standardisation hypothesis must be sought. In particular, re-standardisation as a systematic phenomenon in demographic trends has to be considered more seriously than has so far been the case. Theoretically, it is necessary to supplement the presented approaches with a profound concept of social change which corresponds to general assumptions about the logic of processes of pattern transitions.