The demise of certainty: shifts in aspirations and achievement at the turn of the century

Abstract

The analysis of two British cohort studies (of people born in 1958 or 1970) and one British panel study (of people born in the early 1980s) tracked the educational, employment and marital preferences of three generations of young people between ages 16 and 23/26. It found a steady decline in young people achieving their ambitions. Supporting evidence from in-depth interviews with parents and their children suggested that the perceived need, ability and opportunity to disconnect from tradition and to engage in autonomous decision-making had become the main drivers of aspirations. Although this autonomy was greater for current than previous generations, it remained socially inequitable, with parents and their children accepting the widening gap that separates past and present transitions to adulthood.

Keywords:

• risk society
• uncertainty
• intergenerational comparisons
• cohort studies
• mixed-methods studies

Introduction

The economic crisis of 2008 has reminded the world of the complexity and fragility of inter-connected societies and economies, and the limit to our ability to manage and regulate them effectively. ‘Irrational exuberance’ (Shiller, 2005) brought down a system seemingly stable and well protected because everyone was sharing financial risks that had been sliced and spread between those dealing in it. A catalogue of topical literature, often prepared by those operating inside the financial system, has since demonstrated the ease with which those absorbed into regulatory and banking systems lost sight of the potential downsides of these trades (Auger, 2009; Tett, 2009). As it turned out, what had been deemed calculable risks were far more uncertain than traders, often propped up by advanced mathematical financial models, had imagined (MacKenzie, 2010: Taleb, 2004).

This paper takes a broader look at uncertainty in the late modern British society, focusing on young people's transitions into adult life and, specifically, the labour market. We argue that, in the United Kingdom, uncertainty is becoming more widespread, forming a greater part in everyday lives today than in the past. Moreover, uncertainty appears to have become more ‘acceptable’. Current generations of young people expect to be able to ‘experiment’ with different choices, such as career options, relying on the accessibility and flexibility of the British educational system and labour market to provide multiple (re-)entry opportunities – both to sample options and to return to the mainstream following experimentation. Rarely, however, do they reflect on the possible downsides and long-term costs of these experiments that remain largely unexplored, also by the social sciences.

The evidence presented in this paper puts to an empirical test the core of the ‘risk society’ thesis, developed by Ulrich Beck (1992), Beck and Beck Gernsheim (2002) and others (for example, Bonss, 1995; Giddens, 1994), which posits a major social change to a late modernity. In risk society, changes in institutional structures governing employment, marriage and family life generate more diverse options on how people live their lives. Risk society is further shaped by changes at the level of ideas, which undermine received authority including that of scientific expertise and of bureaucratic and official structures (Lyotard, 1984; Wynne, 1996).

Fostering these trends, globalisation is forcing societies into diverse directions. On the one hand, the growing interconnectedness of markets, the growth of communication networks and the blurring of national boundaries undermine inherited identities and conventions. On the other hand, the lethargy of societal and organisational structures restricts the revolutionary power of globalisation. These limits push individuals who are trying to question, escape and overcome structures of the past back into prevailing social frames (Mills et al., 2008). As vividly demonstrated by the research of the Globalife project, late modern society has created new, frequently ‘flexible’ forms of social and economic life, but also maintained many of society's traditional inequalities (Blossfeld & Hofmeister, 2005; Blossfeld, Buchholz, Hofacker, & Kolb 2011).

In these conditions, social life becomes more uncertain, because people are faced with seemingly greater opportunities to make purposeful choices in more flexible social and working lives. These shifts result in greater individualisation, the process whereby the traditional framework of support that structures the individual's life-course weakens, and ‘your life becomes in principle a risky venture’ (Beck & Beck Gernsheim, 2002, p. 47). The increased diversity of lived experiences does not inevitably reflect greater choice, a growing ability to select from options to choose the one that is most desired. Choices remain structured from the outset as they are limited by what is perceived to be personally achievable or, indeed, socially acceptable; and structured in terms of outcomes as societal barriers continue to define the extent to which social trajectories are permitted to deviate from established paths. Thus, uncertainty also carries the risk of disappointment, of failure to achieve ambitions because the resources and capabilities required to do so remain socially unequally distributed (cf. Mythen, 2005; Savage, 2000; Breen & Goldthorpe, 1999).

The structural uncertainty that defined late modernity changes the parameters within which decision-making takes place and within which personal biographies are crafted. In this world, what is considered to be ‘risky’, as opposed to necessary or inevitable, changes. Tulloch and Lupton (2003, p. 106) demonstrate this duality with respect to the unequal and accelerated economic trajectory of the ‘“risk modern” city’. This environment forces reflection on the ambiguity of the comparatively ordered modernity of biographical histories up to the present and the fragmented post-modern future. Yet despite the emerging ‘new’, the personal-biographical accounts of Tulloch and Lupton's subjects typically remained rooted in concepts of the past, in concerns and perceptions of risk (and crime, specifically), perhaps clambering for old securities.

The present research locates the perceptual and the lived difference of risk and uncertainty rooted in social structures. Liberal welfare systems and economies, such as that of Britain, are particularly open and, arguably, vulnerable to inequitable forces of globalisation. Seeking to harness and exploit the economic force of increasingly un-manageable market forces, they provide little ‘guidance’ to those living within this whirlwind of change that, for many quite literally, melted ‘all that is solid into air’ (Berman, 2010). The destruction caused by these forces is met with individual reaction, sometimes resistance, although it is not clear whether seemingly ‘rational’ responses are calculated, deliberate and reasoned, or passive-reactive and suffered. The analysis of Globalife was based on the scrutiny of national longitudinal datasets and demonstrated increasing uncertainty. Adopting Elster (1989), the project also found evidence of rational decision-making that fitted actions to beliefs, beliefs to evidence, but also evidence to beliefs. Within this framework, the greater uncertainty experienced during transitions from youth to adulthood, which were considered by the Globalife project and are the focus of the present study, has profound repercussions for young people's life-choices, as extended education, delayed marriage and family formation, and flexible partnerships replacing firm relationships confirm (Blossfeld & Hofmeister, 2005).

Our study also encountered an intergenerational chasm in the lived experience of uncertainty across and between generations. We present evidence of the growing uncertainty experienced by young people in the United Kingdom during their transitions to adulthood in the last four decades, and how this growing uncertainty changed the dominant perception of individual responsibility for self-management. Using data from UK cohort and panel studies, we estimate the propensity of young people ‘constructing their biographies’; that is, achieving their aspirations and ambitions, specifically with respect to the preferred educational paths, their employment choices and their partnership formation aspirations. As we are able to use studies that have followed young people born in different decades, we can compare the aspirations and the outcomes, and observe how the relationships between the two changed over time. The evidence is clear: young people have been finding it increasingly difficult to turn their aspirations into a lived reality. Young people increasingly face disappointment and the need to adjust their lives and their aspirations to different, perhaps more achievable, conditions and objectives.

Two studies

We assemble two pieces of empirical research that were conducted successively, using qualitative (cf. Nyhagen-Predelli & Cebulla, 2011) and quantitative (cf. Taylor-Gooby & Cebulla, 2010) evidence. The present contribution updates and integrates the findings of these two studies. It shows how these explorations, conducted independently and using different methodologies, reached similar and compatible conclusions. Both provide empirical evidence of uncertainty in the United Kingdom growing over time and between generations.

The quantitative element analysed two cohort studies and one panel study, exploring the propensity of three generations of young people to realise their personal aspirations with respect to education, employment and partnership formation (marriage). This part of the study sought to observe and to ‘measure’ uncertainty, comparing present with past experiences. To that we added evidence obtained from qualitative research that explored the evolution of risk perceptions and risk-related behaviours comparing the employment choices of two generations of the same families at the time of their entry into the labour market. That part of the study set out to understand the nature and consequences of, and responses to, uncertainty. Here, the qualitative research found ample evidence of a greater openness towards risk-taking and ‘experimentation’ among younger generations, driven above all by the desire for independence, for satisfying curiosity, for exploration and, on occasion, for adventure. Both the quantitative and the qualitative evidence linked uncertainty to breaking with tradition.

The following section briefly describes the studies' content and how their data were used in the present joint analysis.

Measuring uncertainty quantitatively

To explore and measure the rise of uncertainty, we scrutinised three UK datasets, namely:

1.	The National Child Development Study (NCDS), which tracked children born between 3 and 9 March 1958;
2.	The British Cohort Study (BCS70), which tracked children born between 5 and 11 April 1970; and
3.	The British Household Panel Survey (BHPS), which first interviewed a representative sample of the population of British households in 1991, from 1994 has included a separate questionnaire directed at young people aged 11–15.

All three studies asked participants, at the time they were aged 15 (BHPS) or 16 (NCDS, BCS70), about their aspirations for:

i.	further or higher education,
ii.	their preferred work after completing education, and
iii.	whether and at what age they would like to get married.

These three aspirations, shared across the three surveys, formed the basis for the first study, which investigated the change in prevailing uncertainty. As participants were re-interviewed in subsequent years, it was possible to observe whether these young people had achieved their aspirations. To do so consistently, we selected the period when the young people, now growing and grown up, had reached the age of 23 or, in the case of the BCS70, 26.1

In addition to matching aspirational questions (and our ability to record their realisation or otherwise), we also needed to consider whether the time periods covered by the data were suitable for testing the risk society hypotheses of increased uncertainty. Beck and Beck-Gernsheim (2002, p. 48) refer to the individualisation process ‘reaching back to the 1970s and 1980s’ and sharpened by ‘the precarious conditions of a capitalism without work’ in the early 1990s. Giddens characterises risk society as emerging in the ‘late twentieth century’ (1994, p. 5). The data used in the study covered the period between the mid-1970s and early 2000s, and therefore offered an adequate coverage of the key time span (Table 1).

Table 1 Time periods covered, by data source.

Main respondents' age	Information	NCDS	BHPS	BCS70
0 (birth year)	Start of observation; socio-demographic baseline data from parents	1958	1979–1983	1970
15/16	Data about young person's aspirations and relevant parental preferences were collected.	1974	1995–1998	1986
23^a/26	Data about outcomes collected	1981	2002–2005	1996
Note: ^aNCDS and BHPS.

Focusing the analyses on young people who had participated in all relevant sweeps of the cohort studies or in all relevant waves of the panel survey allowed us to record education, employment and partnering choices accurately. However, in the case of BHPS, this had the unfortunate effect of reducing case numbers markedly. The small number of young persons aged 15/16 in the BHPS made it necessary to combine the data from a number of years: we recorded the aspirations of cohorts of young people in BHPS reported in 1994, 1995, 1996, and 1997 when they were aged 16; and the outcomes when the same young people were aged 23 in 2003, 2004, 2005 and 2006. The aggregation of these cohorts provided the analysis with 291 cases of young people in the BHPS sample who participated in all waves and who provided information on outcomes as well as aspirations with respect to all three indicators. The relatively modest sample size warrants a degree of caution when interpreting the results. The samples were larger for the other surveys: 7362 young people in the NCDS, and 1616 in the BCS (see later Table 2).

Table 2 Number of achieved aspirations (uncertainty), by survey.

Aspirations	NCDS (%)	BHPS (%)	BCS70 (%)
0	2.3	4.5	4.5
1	20.6	34.4	26.7
2	46.1	47.8	44.2
3	31.1	13.4	24.6
n	7362	291	1616

In this contribution, we focus the reporting of analysis to NCDS and BHPS. This is because BCS70 proved a less complete source of information due to problems that had affected the survey during data collection. However, it is worth pointing out that, where we were able to include BCS70 in the analysis, it confirmed and complemented the findings of the NCDS and BHPS analyses.

Exploring inter-generational difference in uncertainty qualitatively

The second component of the research was a series of intergenerational in-depth interviews conducted separately with mothers and daughters, and with fathers and sons in 29 families in the English mid-counties (Nyhagen-Predelli & Cebulla, 2011). The interviews captured a diverse socio-economic mix of families, including families in which parents and children shared similar occupations, or where the child had achieved a ‘higher’ occupational grade than their parent (the reverse case was not encountered).2 Parents were aged between 38 and 80, with most being in their sixties and seventies. Their children were aged 17–53 years, with most being in their twenties, thirties and forties. The parent generation thus experienced their education-to-work transitions during the immediate post-war period and into the 1960s, whereas the sibling generation's transitions took place in the last three decades of the twentieth century.

The interviews were conducted in 2004 and 2005, and focused on a comparison of education-to-work transitions and occupational decision-making processes between the generations. The qualitative study also explored the transmission of perceptions and responses to uncertainty between two generations of the same family. Although the topic of the conversation was ‘risk’, the term was, in fact, not used in order to avoid directing or biasing the dialogue (cf. Henwood, Pidgeon, Sarre, Simmons, & Smith, 2008). All interviews were transcribed and analysed manually as well as using qualitative analysis software (NUD*IST). The parents' and their children's narrative account of their own and respective other's biographies were analysed. Specifically, the analysis looked for patterns in the presentations of options, choices and decisions that allowed differentiation between degrees of reflexivity and engagement in the construction of personal biographies.

Constructing key indicators

In the quantitative analysis, uncertainty was measured with respect to three life-decisions that young people need to take: the decision whether to remain in education after completing compulsory education, the preferred job choice, and their preference for forming a partnership, measured in terms of the preference for married life. In each case, we calculated the match between aspiration and outcome, independent of the nature of the aspiration (e.g. whether a young person preferred to marry or not to marry). All statistics that we estimated were based on the balanced panel data; that is, included only young people who participated in the respective surveys at both age 15/16 and age 23. The following measures were used to construct the match indicators.

School-leaving age

The reference school-leaving age was fixed at 16, the final year of compulsory education in the United Kingdom. Using this threshold, for each person, relevant binary indicators were calculated of the expectations as expressed at age 15/16 (young people were asked whether they wanted to stay in education or not), and the outcomes at age 23 (i.e. whether the young person had actually left school by age 16, or whether they stayed on). The cohort studies recorded this information in life-history sections. For the panel study, we extracted the information from subsequent survey waves.

Occupation

In order to compare aspirations and their outcomes across the two cohort studies and the panel study, occupational categories were aggregated since classifications varied between the three surveys. This left the analysis with just three occupational types: manual, non-manual and advanced non-manual occupations. Advanced non-manual occupations included managerial, professional, and associate professional and technical occupations. Clerical and secretarial jobs, personal and protective services, and sales occupations made up the remaining non-manual (service) occupations. Craft and related occupations, plant and machine operators and other elementary occupations were aggregated in the manual occupational category. We only recorded jobs that constituted the young person's main economic activity, thus excluding summer jobs or jobs taken to support other activities, such as further education.

Age of marriage

In the case of young people's partnering expectations, a binary indicator was created to indicate whether a person wanted to get married by the age of 25. We had to select this age because, unlike other surveys, in the NCDS marriage aspirations were recorded in age ranges. The latest age that young people could suggest for their marriage aspiration, and whose outcome could still be observed, was the age of 25.3

Owing to the timing of the next round of NCDS cohort interviews, it was only possible to establish whether young people of that cohort had got married by the age of 23 (rather than the aspiration age of 25). This discrepancy in the aspired and actual age made it likely that meeting the aspiration of marriage was under-recorded. Similarly, the data constraints may have under-recorded the incidence of marriage among young people who had not expected or aspired to be married by the age of 25. In either case, however, differences in trends between generations were sufficiently large to suggest that more accurate data would not have fundamentally altered the general conclusions drawn from this panel analysis.

Education, jobs and marriage across generations

In this section we outline the main statistical findings from our calculations for the matches between aspirations and outcomes across the generations, drawing exclusively on the quantitative analysis of the cohort and panel studies.

School-leaving age

The examination of the NCDS and BHPS data confirmed the inter-generational trend towards more widespread and longer education. Thus, the proportion of young people who intended to leave school at 16 decreased from 67% among those born in 1958 to 14% among those born in the late 1980s. The proportion of young people who did leave school at 16 also decreased from 71% to 16%. This said, fewer young people actually achieved their original objective, be it either leaving school at 16 or staying on after. Overall, 91% of young people born in 1958 achieved their objective, compared with 84% of young people born in the late 1980s. This decline in the proportion of young people achieving their objective was strongly affected by the general decline in early school-leaving. Whereas young people who wanted to leave school at age 16 were almost certain to do so if they were born in 1958, they were just as likely to do so as they were to stay on at school if they were born in the late 1980s. The odds4 of leaving school at 16 among those who wanted to leave school at 16 changed from 31:1 (a near certainty) among the older generation to 1.1:1 (just a little higher than even odds) among the younger generation. The odds of not leaving school at age 16 among those who did not want to leave school at age changed from 4:1 among the older to 8:1 among the younger generation, indicating a stronger relationship in later years.

Current/last job

The decline in the proportion of young people achieving their original objectives was even more marked with respect to their job outcomes. Among the 1958 generation, 37% and 36% of young people in employment at age 23 had aspired to advanced or other non-manual ‘first jobs’ seven years earlier. In the end, 21% and 44% of these young adults worked in advanced non-manual or other non-manual occupations.

Among the late-1980s generation, 55% and 25% of young people working at age 23 had, respectively, aspired to working in advanced or other non-manual occupations. When they had reached the age of 23, 25% worked in advanced non-manual occupations and a further 54% worked in other non-manual occupations.5

Across all aspirations and outcomes, 60% of the generation of 1958 but only 39% of the late-1980s generation had achieved their job objectives at the time they had turned 23. Inevitably, this decrease in the match between job aspiration and outcome reflected the more ambitious job preferences of the younger generation, which required longer periods of post-compulsory education. Already among the 1958 generation, the match between job aspiration and outcome was lowest for those aspiring to advanced non-manual jobs: 31% of those who had aspired to a job of this type also obtained one. However, the late-1980s generation experienced the same level of match between aspiration and outcome: again, 31% of those aspiring to advanced non-manual occupations had obtained them. In fact, the main difference concerned manual occupations: whereas 83% of the 1958 generation with that aspiration did, by age 23, acquire regular employment in a manual occupation, this was only true for 30% of young people of the later generation. In other words, the younger generation experienced that aspirations for both advanced non-manual and manual jobs had become equally unlikely to be met. Expressed again in odds, the odds of achieving one's aspiration for an advanced non-manual job decreased from 0.6:1 for someone born in 1958 to 0.4:1 for someone born in the late 1980s, while the odds of achieving a manual job decreased from 2.9:1 to 0.4:1.6

The decreasing match between aspiration and outcomes was therefore not just a reflection of rising expectations and educational demands, but also of declining opportunities, even at the broad level at which our occupational categories had been defined.

Marriage

The match rate between marriage aspirations and outcomes declined slightly from 53% among the 1958 generation to 46% among the late1980s generation, largely through a combination of changing preferences and changing odds of achieving these. Thus, among the older generation, 89% of young people had aspired to be married by the age of 25, falling to 56% among the younger generations. By the time they had turned 23, 49% of the older generation but only 6% of the younger generation had achieved that aspiration. The main shift occurred with respect to non-marriage. On the one hand, young people who did not want to get married were much more likely to achieve that ambition if they were born in the late 1980s rather than the 1950s. The odds of non-marriage among those who did not want to marry increased from 2:1 among the 1958 generation to 19:1 among the younger generation. On the other hand, the younger generation experienced the ‘collapse’ of their marriage aspiration: the odds of being married by age 23 among those aspiring to be married by age 25 decreased from even odds to fairly improbable odds of 0.01:1.7

Uncertainty: prevalence and features

We pooled the separate evidence of the aspiration–outcome matches to create an indicator of uncertainty, which we defined as the propensity of a young person meeting one, two or all three of their original aspirations. Summing and comparing aspirations with outcomes, the analysis highlights the marked social changes across the generations, as measured by comparing the aspirations expressed by young people with the ‘achievement levels’ corresponding to those aspirations.

Across the generations of young people, the evidence highlighted a steady increase in the propensity of young people not achieving their aspirations at age 23 (Table 2). Thus, whereas the generation born in 1958 had a one-in-three chance of achieving all three of their aspirations, this reduced to one-in-six of the generation born in the late 1980s.8 At the same time, the proportion of young people not meeting any one of their three aspirations almost doubled from 2.3% to 4.5% – a percentage figure also found in the BCS for the generation born in 1970.9

Who managed to reduce uncertainty?

Logistic regression analysis was conducted to compare those young people from our 1958 and late1980s generations who had achieved at least two of their aspirations with those who had achieved one or none of them (Table 3). The analysis included a number of background characteristics of young people and their families that could be expected to affect the propensity of achieving aspirations: demographic information (gender, ethnicity), parental occupation (coded in the same way as for the children), family situation (household type), and the young person's experiences in the labour market (age when started working; the number of jobs to date). Original aspirations were also included in the model to account for inter-generational shifts in preferences. Although the small case numbers available for analysis of the late1980s generation reduced the power of our analysis and thus generated fewer statistically significant results, we still gained insight into some of the drivers that shaped the increase in uncertainty experienced by young people.

Table 3 Logistic regression results of propensity to achieve two or more aspirations (uncertainty).

		NCDS		BHPS
		Odds	Sig.	Odds	Sig.
SOC of father	(Advanced non-manual)
	Non-manual	0.95		0.67
	Manual	0.94		0.77
SOC of mother	(Advanced non-manual)
	Non-manual	1.10		2.03
	Manual	1.17		0.88
	Looking after the home	1.08		2.20
Gender	(Male)
	Female	1.65	***	1.65
Ethnicity	(White)
	Other	0.65	(*)	3.39
Household type	(one or two children aged 16 or younger)
	Single parent	0.89		57.20	*
	3+children aged 16 or younger	0.90		1.42
Number of jobs ever had	(1)
	2	0.70	***	0.77
	3	0.60	***	0.76
	4+	0.52	***	1.28
Age when had first job	(16 or earlier)
	17	0.86		1.41
	18	1.25	(*)	3.87
	19+	1.82	***	5.25
Child's expectations for first job	(Non-manual)
	Advanced non-manual	0.28	***	0.25	**
	Manual	1.44	**	0.34	(*)
Child wants to leave school at 16	(No)
	Yes	3.02	***	0.08	***
Child wants to be married by age 25	(No)
	Yes	0.61	***	0.03	***
Pseudo R ^2		0.14		0.41
Note: SOC = Standard Occupational Classification. Significance: *p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001.

Among the 1958 generation, women were more likely to achieve their aspirations (and a greater number of them) than men, as were those who did not have their first job until the age of 19 (or later). In contrast, people of non-white ethnic background were marginally less likely than others to achieve all their aspirations. Seeking a manual job and wishing to leave school at age 16 also increased the odds of young people from that generation achieving more than one of their aspirations. A greater number of jobs decreased the odds of meeting aspirations – which highlighted the benefits of getting the aspired job at the first attempt.

Some two decades later, women were no longer statistically significantly more likely than men to achieve two or more of their aspirations, although the statistical odds were identical to those estimated for the older generation. People of non-white ethnic background were now just as likely as others to achieve aspirations.10 Among the late-1980s generation, we found single parents to be more likely to achieve several aspirations. This may reflect the less ambitious nature of these aspirations, although these were controlled for statistically, if, in the light of small case numbers, imperfectly.

Aspiring to advanced non-manual jobs and to leaving job at age 16 both strongly reduced the odds of achieving two or more aspirations. The very same association between uncertainty and advanced non-manual job aspirations had been found among the 1958 generation. In contrast, however, the association between uncertainty and educational aspiration had changed between decades, marking one of two most profound differences in experiences between the generations. Whereas among those born in 1958 aspiring to leaving school early, which most then did do, contributed to greater odds of achieving the majority of aspirations, among those born in the 1980s these odds were reduced.

The other profound difference in experiences between generation concerned the shift in association between uncertainty and job changes. Here the association between achieving aspirations and having had fewer rather than more jobs, which was found to be strongly statistically significant for the 1958 generation but disappeared, if not reversed, for the late-1980s generation. Although the findings for the younger generation lacked statistical significance, we will see later that the qualitative study produced similar evidence. Risk theory asserts the growing importance of proactively self-constructed biographies. Repeatedly changing jobs in order to achieve a better match with one's aspiration may thus be part of the struggle to achieve self-determination.11

Uncertainty and biography

It can be contested whether uncertainty is a new phenomenon, quasi-unique to late modernity. Analysing the accounts of education-to-work transitions of current and past generations of young people, Vickerstaff (2003, p. 194) suggests that, contrary to dominant risk discourses, these transitions may have become ‘choppier’ for current generations, but they certainly were not all smooth and trouble-free in the past. The in-depth qualitative study (cf. Nyhagen-Predelli & Cebulla, 2011) confirmed these findings: the older generation experienced their transitions into working life as more ‘certain’ than current generations did. In particular, the older generation noted and, typically, regretted the loss of the ‘job for life’ that, in their view, so adversely affected the lives of their children today. In contrast, for them the force of tradition, mono-industrial structures, and the shackles of geographical ties all narrowed the choice and opportunities for straying too far off the path defined by their parents' generation. But this imposed structure offered security in work and stability beyond. For some among the older generation, of course, their parents' background did offer more choice and opportunity to construct innovative biographies than was the case for others who had inherited less economic and human capital.

Changing self-perceptions and rising autonomy

A key feature of changing relationships between parents and the offspring that the interviews identified was the diminishing parental influence on their children's employment or career decisions – and a concurrent rise in the influence of institutions as mediators and brokers. Parents and their children agreed that there had been a clear break in how younger generations relied on their parents in negotiating their transition into the labour market. Between the 1970s and 1990s, transitions were increasingly being negotiated via public institutions, ranging from advice received at school to that provided by careers advice centres. The detail of that advice went beyond the type of pure brokering service that labour exchanges previously provided to jobseekers among the parent generation, who also continued to benefit from familial connections providing access to work opportunities. But as the brokering by centres of old was replaced by the more hands-off job-advertising and advice service of modern centres in the 1980s, the young generation experienced these changes as opportunities – as well as demands – for more pro-active career development. Inevitably, this led to friction between the older and the younger generations who found it more difficult to understand the other's past and present choices and behaviours. Here the emotional closeness of the parent–offspring relationship appeared to make a key difference to the extent and risk to which this ‘perceptual distance’ turned to conflict.

At a conceptual level, the parent–child generations described their labour market transitions – and their perceptions of the other generation's transitions – in terms of their (assessment of their) capacity to independently and successfully make informed decisions. In broad terms, the qualitative study identified three types of models of decision-making and action (cf. Nyhagen-Predelli & Cebulla, 2011). A comparison of these categories and their prevalence is illustrative of the inter-generational shift in the perception and management of uncertainty.

‘Planners’ were the first category. They shared a decidedly individualistic outlook on their life and what they would be able to achieve occupationally and professionally. This group of individuals talked about planning their various steps, about being fairly in control of their actions and their decisions, if not their outcomes. Importantly, this group considered choices and, in so far as they perceived them as uncertainties, as opportunities for them to proactively construct their biographies.

This group contrasted most clearly with those mothers, fathers, son and daughters who preferred to ‘play it safe’. They typically perceived their transitions options as heavily constrained by local opportunity or by personal and familial circumstances. Unlike the planners, this group often referred to friends and family as sources for guidance, advice or role-models. Frequently paramount in their considerations was the perceived need for security of income, which they rated higher than diversity, options or choice that were more highly rated by our planners.

The final group was labelled ‘surfers’. Their self-portrayal suggested flexible independence of action and decision-making, paired with a sense of ‘going with the flow’. These individuals were highly sensitised to the opportunities that an evolving society offered to them (and others). Their accounts of education-to-work transitions merged personal aspirations with a readiness to diverge from these as new, alternative and interesting options emerged. This group displayed a distinct lack of concern about the future.

Intergenerational shifts

Although these categories are ideal types that could overlap and merge in some of the accounts, such merging and overlapping often reflected the life-changes that the subjects reported and echoed the complexity of the post-hoc accounts and current lived experiences. In most of the biographical stories, however, the singularly planning, safe-playing or surfing narratives were dominant. Striking was the extent to which the balance in the distribution of these three ideal types shifted between the generations. Whereas among the parent generation 18 of the 29 individuals' account corresponded to that of the safe-player, seven to that of the surfer and just four to that of a planner, in contrast only eight of the offspring generation could be described as safe-players, while another eight were planners and 13 were surfers (Table 4).

Table 4 Risk types (qualitative study): generational patterns.

	Parents	Offspring
Planners	4	8
Safe-Players	18	8
Surfers	7	13
All	29	29

Thirteen of the 29 parent–sibling pairs displayed matching risk orientations and behaviours, either because both generations were planners (two cases), safe-players (eight cases), or surfers (three cases) (Table 5).

Table 5 Risk types (qualitative study): shifts between generations.

	Offspring
Parents	Planners	Safe-Players	Surfers	Total
Planners	2		2	4
Safe-Players	2	8	8	18
Surfers	4		3	7
Total	8	8	13	29

Not only the numerical distribution changed between generations: the underlying social construct and narrative account that they represented also changed. The older generation's accounts of their transitions and their reflection on their offspring's transitions 20 or so years later remained deeply rooted in, and tied to, locality and an awareness of shared, if changing ‘constructural’ constraints.

These constraints were structural in the sense that locality was seen to continue to limit options for both generations, if less so for the current generation. The constraints were concurrently constructed in the sense that both generations regarded the choices they had as bounded by expectations – typically those passed on by the previous generation of parents, but also through the dominant social culture in the locality. Yet despite these constraints, the older generation often described their transitions as the result of independent, purposeful actions, if building, unquestioningly, on local and familial resources. For this generation, the main difference to the construction of their education-to-work transitions was the radius of the setting within which these decisions were taken. Geography, industry and occupations were experienced as stratifying one's choices more deeply than was the case for younger generations.

This account closely matched that of the younger generations, if observed from the reverse perspective. Here the loosening of geographical, industrial and occupational ties inherited in the locality and through family and class were experienced as liberating and, at the same time, challenging. They were liberating because they seemed to offer more freedom and greater fluidity, opportunities for ‘reflexivity in action’, as the younger generation readied itself for, but also embraced, the need for change. But this was also the core of their challenge. This generation needed to plan their lives, but felt increasingly disconnected from the world and experiences of their parent generation that could no longer provide guidance or support from established informal networks. None of the younger generation obtained their first job with the help of their mother or father. Most of the older generation could, at least, recall one member of their extended family inspiring their own first-job choices or, indeed, helping them obtain that first job.

For the older generation, breaking with these traditions was out of the question. It would have had adverse consequences for how their family and social environment would have perceived them. It might also have entailed financial penalties. None of this was the case – or a concern – for the younger generation. Materially independent, the belief of being in charge, and the perception of formal security detached from custom had made these worries redundant for the current generation of young adults.

Unlike the previous generation, the younger generation constructed their education-to-work transitions as medium-term career projects, involving changing or planning to change jobs, considering greater physical mobility – and along with it, the stretching of familial ties across larger geographies. These ‘planning constructs’ were present in the accounts of people in all occupations, but with a difference in their chronology. Those equipped with more human and educational capital used planning as part and parcel of their development process, of their testing options before settling for a preference (largely surfers). Some, but by no means all, then also sought to apply the same experimental approach to progress to more desirable positions. In contrast, those equipped with less educational capital and eventually working in occupations that, measured in terms of earnings capacity, society valued and rewarded less generously, (re-)planned their careers after they had already decided on, and taken up, an initially preferred occupation and job. The former thus practised reflexivity to design their career trajectories; the others practised reflexivity to correct their trajectories. This differentiation was remarkable by its clarity with which the younger generation told their stories – and by its absence in the accounts of the older generation.

The older and the younger generations' accounts mapped neatly onto the findings of the quantitative analysis. As seen earlier, that analysis suggested a positive link between greater job mobility and meeting aspirations; that is, dealing with uncertainty. The analysis found this pattern for the late-1980s generation, but not the 1958 generation, for whom uncertainty was most reduced if they had obtained their aspired job with their first position – and did not change jobs frequently thereafter.

Between generations, responses to uncertainty have become disconnected from the perceived need for creating personal and familial security, and connected to perceptions of aspirations and judgements as to one's capability of meeting these aspirations. Coping with uncertainty was increasingly argued to be an individual pursuit. This did not imply a severing of parent–sibling relationships, although some accounts suggested families had come close to that point when disagreeing over chosen career paths. Siblings continued to consult and seek the advice of their parents, but they increasingly did not feel that the advice was relevant. This openly acknowledged clash did not threaten relationships (as yet), because the parents reciprocated, typically arguing they lacked the understanding of late-modernity's complex social and economic functions and mechanisms to be able to provide effective and robust advice. Mutual tolerance, if not respect, for choices was the product of, occasionally, deliberate negotiations and, sometimes, tacit acceptance. Parents who felt they had been pushed into careers they would not have chosen themselves tended to perceive their children's greater choice as a welcomed social development. For those few, acceptance of their children's choices turned to encouragement.

Conclusions and discussion

The integration of the two studies reported in this article provided complementary empirical evidence of an increasing incidence of uncertainty in the United Kingdom in the last three or four decades, reflecting the main socio-economic transformations that have taken place in education, in the labour market and in personal relationships. These changes have not only extended adolescence and prolonged the transition periods into functional independent adulthood.

The findings of this research echo those of the Globalife project. Moreover, the qualitative research provided empirical confirmation of the thesis by Blossfeld and Hofmeister (2005) that, in the face of growing uncertainty, young adults strive to retain rational decision-making, plotting their careers, conscious of and weighing their options, correcting decisions and pathways as new opportunities and needs emerge. This said, sustained inequalities to navigate uncertainty remained evident as lower (working)-class men and women found the scope for and scale of alternative pathways curtailed by structural opportunity or personal aspiration.

Uncertainty has changed the probability with which aspirations, previously deeply rooted in traditional concepts of learning, working and marriage, can be achieved as they had originally been intended. Although aspirations had become more and more detached from concepts rooted in tradition, family and place, for many these adjustments were outpaced by the speed with which social realities changed and new social risks emerged (Bauman, 2007; Taylor-Gooby, 2004). The result was a gradual increase in uncertainty across generations.

The disconnection of personal aspirations from traditional expectations was replaced by a growing desire for self-determination, whose probability of realisation was curtailed by structure and construction. Opportunity structures remain a major obstacle to self-realisation, despite, at least in the realm of the labour market, evidence that the younger generations employed more experimentation in order to break down these barriers.

For those from a lower/working-class background, however, opportunities for this experimentation were much more limited, while current costs and future risks were potentially higher. Whereas more economically advantaged young adults perceived experimentation as a part of their personal development, for others the uncertainty of gaining a return on investment by deviating from a traditional pathway on the transition into adulthood reduced their openness to this type of risk-taking. As Breen and Goldthorpe (1997, 1999) have argued, investment in education is a much more costly endeavour for those who – and whose families – depend on wage incomes rather for the better-off salariat. This cost calculation reduced the likelihood and the scope for experimentation during the life-course. Hence, the young adults in the qualitative sample still made very rational decisions when they decided for or against deviating from traditional paths of transition to adulthood.

This said, both quantitative and qualitative evidence suggests that more effort is dispensed on self-designing one's education-to-work transitions, although this must not be confused with the strategic planning and implementation of career ambitions. On the contrary, the ‘career’ itself may well have slipped into the background, overshadowed by the desire to demonstrate control over one's life-choices.12 None of this rules out frustration, as Furlong and Cartmel (1997) demonstrated for working-class school children. In fact, the qualitative study encountered little, if any, acknowledgement of the potential costs of experimentation, nor are we aware of any estimations of the cost and benefits from speculative, experimental and delayed career entries of the kind that many of the younger generation were actively trying out or passively experiencing. We simply do not know whether the labour market, ultimately, rewards or punishes experimentation. For cross-generational families, the costs of experimentation can be high, particularly if the experiments themselves receive – or require – parental financial support. More remarkable is, perhaps, the high level of acceptance among the older generation of their children's novel career construction strategies.

Whither certainty? Normalising uncertainty?

The histories and experiences that were told reflected adjustments and adaptation of, and between, generations amidst experimentation (or even upheaval). But there were also accounts of differences and how these could separate generations was it not for a mutual acceptance of the difference. Although the parents and siblings in the qualitative study experienced uncertainty, their life-courses appeared not to have been pluralised or diversified to the extent that the individualisation thesis at the heart of risk theory might proclaim. Whereas the data suggest increased heterogeneity in choices and ambitions, young people, not unlike their parents, sought reference points and anchors for their own decisions in their parents' as well as their peers' experiences. Whereas choices differed in quantity and quality, the basis for deciding a course of action typically reflected a rationalised concern for a career path that remained attached to a historical legacy created by previous generations. Decisions of this kind were structurally dependent and channelled by social expectations. Viewed from a distance that allows us to capture the macrocosm beyond the microcosm, ‘the majority of individuals in fact opt for the “default” standard life course’ (Mills, 2007, p. 76). The ‘default option’, once innovative, exceptional and non-traditional, thus becomes the norm. The inter-generational interviews have thus encountered an acceptance of change among both generations. This acceptance entailed a generalisation, which must be affected by societal change, of how either generation perceives and interprets the experiences and experiments of the other. This realisation is effectively a concession that negotiates situations of potential conflict arising from difference.

Acknowledgements

The two studies upon which the current paper is based were both funded by the Economic and Social Research Council (under the Social Contexts and Responses to Risk programme RES-336-25-0001, and research grant RES-000-22-1751). We would like to thank the ESRC for its support. In addition, we would like to thank the numerous researchers who contributed to the studies, including Professor Peter Taylor-Gooby (University of Kent), Dr Noel Smith (now University Campus Suffolk), Abigail Davies (Centre for Research in Social Policy, CRSP, Loughborough University) and Lyn Cox (formerly CRSP). In particular, we would like to express our gratitude to Dr Line Nyhagen-Predelli (Loughborough University) for her critical contribution to analysing and interpreting the qualitative data.

Notes

^1. Unlike the BHPS, which is an annual longitudinal survey, cohort studies re-interviewed participants after a gap of a few years, which made it impossible to achieve an exact age match.

^2. Sixteen of the parents were or had been manual labourers, 12 (had) worked in low-skilled (eight parents) or medium-skilled (four parents) service occupations, and one in advanced level services. Ten of the offspring generation were manual labourers, while four worked in low-skilled and 15 worked in medium-skilled service occupations.

^3. The next recorded age range was 26–30 years, for which no corresponding outcome data were available at the time when we conducted the study.

^4. The odds are calculated by dividing the percentage of young people achieving their aspiration by the percentage not achieving the aspiration among those expressing a given aspiration.

^5. We also calculated the match between aspirations and first regular jobs. Aspiration–outcome statistics using this indicator were almost identical to those estimated when using the current/last jobs variable.

^6. The odds of achieving an aspired non-manual job (other than advanced non-manual) decreased from 1.8:1 to 1.5:1.

^7. Only 7% of young people born in the 1980s who had aspired to be married by the age of 25 had been married by the time they reached the age of 23; 93% were still unmarried.

^8. The generation born in 1970 occupied a mid-range position with one-in-four achieving all three of the aspirations by the time they reached the age of 26. The choice of the ‘outcome age’ was again dictated by the data.

^9. Societal changes may dictate that later generations take more time to construct their biographies than earlier generations. Choices at age 16 and outcomes at age 23/25 may therefore not be final. The current research should hence be repeated with longitudinal data that, when available, captures matching later life-stages.

^10. The odds even suggested a greater propensity people of non-white ethnic background to achieve aspirations, although the statistic proved not significant.

^11. A separate study, not further reported here, found that young people of the late-1980s generation living in deprived areas used multiple job changes to the greatest effect to significantly improve their job satisfaction. Their job satisfaction tended to be lowest when their first job had been the type they had aspired to at age 15.

^12. Note that in a separate analysis of job satisfaction among our generational cohorts we found a statistically significant link to achieving higher occupational status only for the 1958 generation, but not for the late-1980s generation.