The Impact of the Economic Crisis on Youth Trajectories

This article analyzes the impact of the economic crisis on the patterns of transition followed by Catalan young people. In particular, it does so by examining to what extent the crisis has affected the extension, de-linearization, reversibility and diversification of their trajectories. These processes have been detected at the European level and are often linked to a context of greater opportunities. The article focuses on Catalonia, an example of a familistic youth regime. Results show that, in a context of crisis and for the Catalan case, transitions take longer, linearity and reversibility increase and, although vulnerability rises, the typical trajectories remain stable. Thus, risk avoidance, mostly through family support, has become even a more dominant strategy than during the expansive economic period. This reinforcement of the traditional pattern of transition, in which the family of origin has a central role, is expected to strengthen social reproduction.

• The 'extension' of the youth phase was first indicated by Cavalli and Galland (1993). A broad consensus has emerged that this phase increased in length over the 20th century and that this extension has accelerated in recent decades. The aforementioned authors have conceptualized youth as a phase of life more than as a transitional moment; others even defend the arrival of a new period of life-for example, Arnett's 'emerging adulthood' (2000). The extension of youth is related to the extension of the sub-transitions that form it: the increase of the level of instruction of the new generations linked to the extension of the educational period; the delay in the labour transition due to the enlargement of the educational period and the increased vulnerability of young people in the labour market; and the delay in leaving home and forming one's own family due to the enlargement of the previous subtransitions and the cultural changes that many authors link to the concept of post-modernity. On the other hand, youth also extends its lower end, that is, the emergence among adolescents of attitudes and behaviours that previously were typical of the youth phase. • The second dimension of change is the rise of 'non-linear transitions'. In Fordist societies, the traditional sequence of transition into adulthood implied a relatively short period of studying; followed by a (again relatively) quick labour market insertion; a short stay at the parental home after that, followed by leaving home, massively through marriage; and shortly afterwards having the first child. It has been argued that this linear pattern was not as common as it is believed, and that the underestimation of non-linear patterns in the past may be due to theoretical and methodological issues. However, there is a wide consensus that non-linear transitions are increasing and have replaced linearity as the dominant pattern (Pollock, 2008). The idea of a change from linear (rigid) to non-linear (flexible) societies can be found in authors such as the aforementioned Bauman, Beck or Giddens; it reached the field of life-course studies (Buchmann, 1989) and is now commonly accepted in youth studies. Today, in a post-Fordist model of transition (Baizán et al., 2002), working before the end of the educational career, lifelong learning or leaving home before achieving occupational stability have become more frequent. • 'Reversibility' refers to the return to a stage, which, in the linear logic, has been overcome, that is, returning to inactivity (from activity) or to the parental home (after having left it). In the traditional patterns of transition, once one phase had been overcome, it was unusual to return to the previous one. The stability characteristic of the Fordist societies facilitated a smooth transition into the labour market; and personal life, in a context of solid values (Bauman, 1992), tended to follow the same linear patterns. In our post-Fordist societies, reversibility of events has grown, to the extent that it can be a dominant characteristic of the youth transition. A yo-yo image (EGRIS, 2001;Machado, 2001) has been used to describe these types of trajectories. Reversibility is one of the mechanisms through which the linearity of transitions has been broken, but it can also occur without changing the traditional sequence of sub-transitions. • 'Diversification' is the last of the considered transformations, which constitutes a general trend in contemporary societies. This issue is closely linked to the general debate on individualization, which also affects the life cycle.
The de-standardization of transitions has been explained as the outcome of the increase of self-reflexive processes in a context of fading of the social constrains (Beck, 1992); however, Furlong and Cartmel (1997) have argued that the de-standardization of trajectories means that now individuals have to face alone the old structural constraints. Leccardi (2005) explains this process as 'biographical subjectivization'. In any case, there is a consensus that, in the past, transition trajectories were relatively short and mostly linear, and therefore the diversity of typical trajectories was reduced. The extension of the youth phase and the rise of non-linear and reversible trajectories have widened the scope of pathways that a young person may follow. In the 'Fordist' model of transition, the types of trajectory were few; today, Stauber and Walther (2002) use the concept of de-standardization to explain the change to a context in which diversity and unpredictability are the norm.
Much of the debate on these changes has focused on the extent to which they result from an improvement of the living conditions and opportunities that new generations have. The choice biography concept (Du Bois-Reymond, 1995) emphasizes the self-reflective construction of the young person's biography in a context of greater resources (economic growth and consolidation of social policies). According to this perspective, social inequalities are declining and the trajectory an individual follows is much less predictable than in the past and depends largely on his/her conscious decisions. This has been widely disputed, notably by Furlong and Cartmel (1997), who have pointed out that the persistence of structural constraints are hidden by 'biographical subjectivization' (Leccardi, 2005).
These changes have not occurred with the same intensity throughout Europe, as they depend on the situation and characteristics of each country. Esping-Andersen's (1990) typology of welfare regimes has been used in order to explain the variability in youth transitions across Europe. Blossfeld et al. (2005), Cavalli and Galland (1993), Walther (2006) and Van de Velde (2008) have produced complementary models that show the influence of institutional and cultural factors on the dominant patterns of transition in each country. Van de Velde (2008) highlights the interaction between dominant values and welfare regimes in shaping these patterns.
According to her results, the comprehensive Nordic welfare regime supports actively individualized, experimental and long transitions; the state tends to guarantee a certain independence from coming of age and, through universal and flexible social policies, stimulates a long educational career combined with periods of professional activity. The continental regime has a corporate bias that pushes young people to 'place themselves' within a stratified educational system, which tends to define the future social identity and position of the individual; under this pressure, trajectories tend to be linear and quick and family support (or partial dependency) is legitimated. Transitions in the Anglo-Saxon countries are marked by their liberal culture and welfare regime, which encourages an early and risky autonomy among young people; youth trajectories tend to be shorter and work orientated; leaving home is expected and tends to happen at an early age; and usually, longer educational careers are self-financed, followed by a quick access to paid work and to marital and parental statuses. Finally, the logic of 'installing oneself', characteristic of Southern Europe, results from its labour market segmentation and weak public policies, which strengthen family ties. There, the dominant pattern of (a delayed) transition consists in 'getting a stable job, getting married and buying a house' (Van de Velde, 2008: 62). Lesnard et al. (2010) and Moreno (2012) confirm that the transition to adulthood in Southern Europe remains marked by its historical family systems. Moreno describes a: familism embedded in the culture and a transition regime within a welfare system characterized by limited development of youth emancipation support policies. Thus, intergenerational solidarity and a [family] dependence culture play a much greater role than in Northern European countries as a protective mechanism against the risk of exclusion and poverty. In the case of Spain, the combination of both factors has given rise to a strategy based on negotiation between parents and children for limited autonomy within the parental home. This is reflected in the extended transition to adulthood and the scarce value placed on independence. (Moreno 2012: 42)

Youth Trajectories in Catalonia as a Case Study from Southern Europe
This article analyzes the impact of the economic crisis, through the filter of the familistic 'youth transition regime' (Walther, 2006), on the trajectories followed by Catalan young people in their transition into adulthood. This case study should shed some light on the nature and dynamics of the general trends of change in youth trajectories identified at the European level.
Catalonia is a Spanish region ('Autonomous Community') with a population of 7,546,522 inhabitants, of which 1,795,650 (23.8 per cent) are young people between the ages of 15 and 34 (Continuous Population Census 2013-Catalan Statistical Institute). Catalonia has a high degree of political autonomy, although most of the main policy areas that affect young people (such as the structure of the educational system, educational subsidies, labour market laws and family assistance) depend on the central administration. Catalonia is part of Ferrera's (1996) 'Southern model of welfare', characterized by a highly segmented labour market, where most young people develop the role of outsiders, dealing with high structural unemployment, temporary contracts, low wages, an extended informal economy and, most recently, over-qualification. The scarce occupational opportunities are not balanced by social policies, as social insurance schemes are limited 1 and contributory, which reinforces social reproduction. These structural constraints tend to increase individuals dependency on the resources and support their families can provide. Strong intergenerational ties and social reproduction tend to be the outcomes of this model (Alegre, 2010).
Research on youth trajectories in Catalonia has shown that even during the previous expansive economic cycle, this familistic welfare regime stimulated traditional patterns of transition. This means that 'linear', 'non-reversible' and 'extended' trajectories tended to be dominant. Moreover, only a reduced set of typical trajectories ('limited diversity') was found (Miret et al., 2008;Serracant, 2011).
This restricted impact of the general trends of change in youth trajectories in Catalonia could be explained by the characteristics of its welfare regime: the agebased labour market segmentation and the weakness of social policies renders the family as the key actor in shaping the opportunities a young person has. In the end, family support materializes in the form of a long cohabitation in the parental home that allows young people to achieve university qualifications, occupational stability or financial savings. Thus, the combination of strong family ties, a segmented labour market especially harmful to young people and the weakness of public policies favours a linear logic in which each step is carefully taken.
The current economic crisis has increased the vulnerability of young people in the labour market and weakened the policies and programmes that could offer them alternative resources. In Southern Europe, this has occurred to a much higher degree than in the rest of Europe. In Catalonia, from 2007 to 2013, the unemployment rate has risen from 6.5 per cent to 23.4 per cent, and there are important age-based differences: the unemployment rate has risen from 5.7 per cent to 19.3 per cent among those older than 29 years; and from 9.3 per cent to 37.1 per cent among people aged 16-29 years. Thus, youth unemployment follows a similar pattern than in most European countries (it tends to double the adults' rate); however, the high general unemployment turns into extremely high youth unemployment. Moreover, the length of the crisis has resulted into long-term unemployment: in 2013, 44.0 per cent of the unemployed young people had been in this situation for at least one year, and only 16.8 per cent of them were entitled to unemployment benefits. 2 Thus, despite the fact that housing costs have dropped substantially during the crisis (-22.5 per cent), 3 between 2007 and 2012, the poverty rate among young people has risen from 14.9 per cent to 21.6 per cent (Instituto Nacional de Estadística or National Statistics Institute [INE]).
To what extent has the crisis altered the dominant patterns of transition? There has been little quantitative research on the impact of the current economic crisis on youth trajectories. The economic crisis and the reduced public spending could stimulate a conservative strategy; spread anomic, unstable choices; or simply promote pragmatic adjustments. However, researches carried out both in the United States (US) (Harris, 2010) and Europe (Gentile, 2010) indicate that the crisis tends to reinforce traditional, conservative patterns. 4 Moreover, the differences in the youth welfare regimes, previously explained, suggest that uncertain contexts, where youth vulnerability becomes a structural feature, stimulate traditional patterns of transition.
For the Spanish case, Moreno (2012) finds that in a context of crisis, youth trajectories are massively dependent on both residential and economic support from their families. Gentile's (2010) research focuses on young people's housing strategies (including 'boomerang kids') during the crisis and confirms family dependence. However, it is important to analyze how the crisis has modified youth trajectories, as this could shed some light on the nature of the patterns of change in youth transitions.
Following these previous findings, this research hypothesizes that, for the Catalan case, the economic crisis will have reinforced the traditional pattern of transition into adulthood in Southern Europe. In particular, the research compares the main trajectories of transition between the last economically expansive and current recessionary periods. Specifically, we analyze the extent to which the economic crisis has affected the extension, de-linearization, reversibility and diversification of youth trajectories. Thus, from one period to the other, we expect to find an increase in the 'extension' and 'linearity' of the trajectories and a decrease in the 'diversity' of the main trajectories followed by young people. Finally, we expect to find an increase in the 'reversibility' of youth trajectories, although delimited to the first years of the crisis. This last trend would constitute a temporary change in the traditional pattern. It would be explained by the fact that the crisis may have affected some of the young people that started their transitions before it begun (i.e., losing their jobs and/or having to return to the parental home). Later on, reversibility should decrease as in a linear and extended logic, steps are carefully taken.
The underlying mechanism that would stimulate these changes is an adaptation of young people's (and their families') desires to the reduced opportunities available in a context of crisis. The interaction of desires and opportunities not only impinges on the transition to independent status (Van de Velde, 2008) and strategies adopted to leave the parental home (Jurado, 2001), but on educational and occupational mobility as well (Beller and Hout, 2006).
The operationalization of the indicators used to measure these changes is explained in the results section in order to facilitate comprehension.

Data
This research is primarily based on the Catalan Youth Survey from 2007 and 2012, an official survey of the Catalan regional government and carried out with the support of the Catalan Statistical Institute. The Catalan Youth Survey is fully representative of the Catalan young population. It focuses on youth transitions, although it also offers limited information on other issues, such as, cultural consumption, participation and health. Its most relevant characteristic, for our purpose, is that the questionnaire includes three sets of retrospective questions on educational, labour and housing/family sub-transitions. Introducing retrospective questions in a standard survey is a more affordable and quick method to achieve information to reconstruct life-course trajectories than setting up a panel survey, and attrition is avoided. However, the information gathered is less reliable, as interviewees are asked to recall events that may have happened long ago (Giele and Elder, 1998). To try to minimize this, the questionnaire should ask for 'basic events' in the individual's live. This is what the Catalan Youth Survey does, and the two previous surveys (2002 and 2007) have proved its consistency when used to recreate the basic features of the individual's trajectories. Finally, comparisons between both surveys can only refer to the Catalan population with Spanish nationality, as foreign-born young people were underrepresented in the 2007 sample. The main technical characteristics of both surveys are summarized next in Table 1.
In addition, some complementary information has been extracted from the Catalan Labour Force Survey.

Results: The Impact of the Current Economic Crisis on the Trajectories of Transition
In order to analyze the impact of the economic crisis on youth trajectories, the focus will on the effect it has on the four tendencies identified at the European level (extension, reversibility, non-linearity and diversification). Each of these four patterns of change will be defined and operationalized and results calculated for 2007 (at the peak of economic growth) and 2012 (when the crisis was well advanced). The source of data will be the Catalan Youth Survey, except in those cases where there are official standard indicators from the Labour Force Survey.

Extension
As it has been hypothesized, in a context of crisis, and especially in a familistic youth regime, it is expected that trajectories will take longer, as young people will adopt conservative strategies in order to avoid risks. Thus, the traditional pattern of waiting in the parental home and delaying the key sub-transitions should be reinforced.
In order to measure the extension of youth trajectories, four indicators have been selected, each of them referring to a particular sub-transition. The indicators are shown in the first column of Table 2. In this case, the data is split into age groups in order to control the influence of variations in the demographic structure (in 2007, there were more young people aged 15-19 than in 2012, where the 30-34-year-old age group was bigger). 5 The age groups vary depending on the number of cases in the sample.
Results are, with some exceptions, as expected: the economic crisis is extending the youth phase, that is, delaying the achievement of certain goals. This has happened through a delay in the educational and labour transition. Thus, the proportion of young people that study has grown, especially among the younger age groups; and those young people that have achieved and maintain an occupation have YOUNG 23 (1) decreased dramatically. However, these changes have not led to an important delay in the housing transition. This could be explained by the fact that young people living outside the parental home were already a minority before the crisis, especially in the age group 16-24 years. In other words, in a Southern European context, this age group tends to remain at the parental home regardless of their activity status (be it student, unemployed or employed). Complementary, it has to be considered that housing costs have dropped substantially during this period, 9 which may have diminished the drop in the proportion of those who live outside the parental home. Finally, the data show a minimum increase in the proportion of young people with children.

Reversibility
As in the case of extension, the change from a context of economic growth to one of crisis is expected to increase reversibility, especially during the first years. After a certain period, it is expected that a conservative strategy of a delayed transition might reduce reversibility. In this research, reversibility is considered to occur when, in a particular subtransition, the individual returns to a previous stage. For example, with regard to the educational sub-transition, reversibility occurs among those young people that had ended their studies only to subsequently return to school. In this article, in order to limit the analysis to the expansion and recession periods, reversibility refers only to those individuals that (apparently) finished the sub-transition (i.e., stopped studying) and returned to it during each of these periods. The indicators selected to measure reversibility can be seen in the first column of Table 3.
Results are as expected, except for the labour transition. With regard to the educational sub-transition, those young people that stopped studying and returned to the educational system rose from 22.5 per cent during the expansive period to 28.9 per cent during the recession. The percentage is calculated over all the young people that stopped studying in that period. Thus, the extension of the educational career detected in the previous section is not only due to the fact that young people tend to continue their studies without interrupting them but also to an increase in the proportion of young people that return to the educational system after having left it.
The stability in the labour sub-transition is relatively unexpected, as unemployment has increased dramatically. However, the indicator does not refer to the unemployed but only to those that lost their jobs and, in the 2008-12 period, many young people that entered the labour market simply did not find their first job, a situation that the indicator does not include; in addition, in 2007, it was much easier to get a new job after losing the previous one, something much less frequent in 2012; finally, it is possible that some of the young people that in an expansion period would leave their (unsatisfactory) jobs expecting to find a new one, keep them during recession.
The crisis has had a clearer impact on housing reversibility, as the proportion of young people that left the parental home and then returned doubled from one period to the other (from 7.8 per cent to 14.3 per cent). Similarly, the percentage of young people that started living with their partner and stopped doing so also increased in the recessionary period (from 6.7 per cent to 11.2 per cent). This result apparently contradicts the official data on separations and divorces of the Spanish Statistical Institute, which show a decline in these events from 2007 to 2009 (although in 2010, divorces  (1) increased). This decline may be related to the monetary cost of trials and to the risks of living alone (or with children) in a context of economic crisis. However, our indicator not only refers to official separations and divorces among married couples but also comprises all situations in which the young person has stopped living with his/ her partner, including non-married couples. Relationships among young people tend to be less stable, both due to an age effect (often these relationships have just started) and to a generational effect (the new cohorts in Catalonia and Spain tend to split up more frequently). This higher volatility among young people may have increased not only due to a generational trend (as only a five-year span is considered) but also to the tensions generated by higher unemployment and vulnerability.

Non-linearity
De-linearization can be understood both as the outcome of an increased vulnerability and as an innovative strategy through which young people optimize the augmented available resources to tailor their trajectories. In a context of crisis in which family support facilitates the rise of conservative strategies, it is expected to find an increase in traditional linear patterns of transition. Linearity is broken, as operationalized in this article, when one sub-transition (for example, the labour one) starts before or parallel to the sub-transition that, according to the traditional pattern (education-work-leaving home-family formation), precedes it (in this example, the educational one). Table 4 shows, for each of the two considered sub-transitions, the proportion of non-linear trajectories during the periods studied.
Unlike extension and reversibility, non-linearity has decreased during the recession period. With regard to the educational sub-transition, young people that started working (as main activity) before finishing their studies (or at the same time) descended from 68.1 per cent (expansive period) to 45.0 per cent (recession period); the proportion of those who left home before finishing studying has dropped by more than one half (from 36.6 per cent to 14.6 per cent); and there also has been a decrease in those who started living with their partner before the end of their studies (7.7 per cent to 5.7 per cent). The same occurs in the labour sub-transition: those who left the parental home before or parallel to when they starting work (as main activity) went from 23.1 per cent (expansive period) to 8.3 per cent (recession); and those who had their first child before or at the same age as when they starting work also decreased (from 2.4 per cent to 1.6 per cent).

Diversification
With regard to diversification, the economic crisis is expected to reduce the adoption of risks, and it can be hypothesized that experimental and innovative trajectories will decline, while traditional patterns of vulnerability will increase. Therefore, a reduction of the main types of trajectory is expected.
The methodology used in this case consists in building and comparing two typologies of trajectories of transition: one for the expansive period and the other for the recessive period. This is, clearly, an indirect approach as the final number of trajectory types has been decided by the researchers themselves based on statistical and theoretical considerations. Statistical, as the groups were intended not to have extremely different sizes; and theoretical, as each resulting group represented a pattern of transition that had been previously identified or suggested by previous researches (AQU, 2008;Casal, et al., 2004Casal, et al., , 2006. The typologies offer a holistic view on youth transitions as they comprise the three main sub-transitions: ending the main period of formal study; entering the labour market; and leaving home/forming a family. Thus, the results highlight the principal interactions between these subtransitions. This approach no doubt blurs precise information on each of the considered sub-transitions. However, it generates an integrated view of the different types of basic trajectories followed by young people and it shows their interdependence (Pollock, 2007;Robette, 2010), which facilitates further qualitative research.
The typologies focus on young people aged 30-34 years, as this group has already started or ended most of the three sub-transitions. This integrated analysis of the three sub-transitions has been possible due to the fact that the questionnaire of the Catalan Youth Survey gathers information from when respondents were 15 years old up to their current age (30-34 years). Present-time information would not have been so useful to build up the typology, as it focuses on processes more than on results. Thus, the first criterion to select the variables was that they referred to one of the three considered sub-transitions (variables referring to other life domains were excluded, even if they could be linked to transitional trajectories, like those related to health issues); and the second criterion was that they should offer dynamic information, that is, which refers to the trajectory followed by the young person (the housing/ family status is the only variable that focuses exclusively on present time). Thus, the variables selected or created to build the typologies come from these retrospective questions (for example, 'age at leaving home') or explain part of the trajectory (for example, 'instruction level'). Table A2 in the Appendix shows the variables included in the typologies.
The technique used to build the typologies is a two-step cluster analysis, which allows a quick and efficient treatment of large datasets and the use of categorical and continuous variables. However, all the continuous variables have been converted to categorical ones. In a first phase, the procedure reduces the size of the dataset in a pre-clustering process and groups the data into conglomerates, which form the basis for the next hierarchical conglomerate analysis. There were 574 and 718 cases to build the 2007 and 2012 typologies, respectively. 11 Table 5 shows the main types of trajectory resulting from each typology.
Both typologies coincide in the number of types of trajectories (seven) and, most importantly, in their main characteristics. Briefly, the main change appears with regard to the 2007 trajectory, 'working class and separated from partner'-mostly formed by women with low qualifications that left the parental home at an early age to live with their partner; most persons with this trajectory were unemployed or had a precarious job, and many had separated, so they were clearly a vulnerable group. In 2012, a different trajectory appears, the 'post-compulsory precarious', formed by young people with basic or mid-level qualifications, who have experienced unemployment and/or precariousness. The rest of the six types show striking similarities.
Three types of trajectory follow a linear pattern: the 'working class' trajectories group those young people that abandoned their studies (with compulsory qualifications or lower) to enter the labour market, where they experienced more unemployment and precariousness than the average; despite this experience, they left home earlier than most young people. The 'post-compulsory linear' trajectories follow the same traditional pattern (studying-working-leaving home) but events occur later than in the previous group: they achieved post-compulsory qualifications and had a later entrance into the labour market, where they experienced low unemployment and precariousness and could leave the parental home without difficulties. Finally, the 'university linear' trajectories follow the same relatively comfortable pattern as the previous one but, again, events occur later as these young people achieved university qualifications. Both typologies also detect three similar trajectories marked by a less linear and longer pattern: the 'late university' trajectory is marked by a delay in finishing studies, which led to a late entrance into the labour market and, therefore, a long stay in the parental home. The 'living with parents' trajectories comprise a wide range of young people (from those who left school early to university graduates and includes many different working experiences) united by the fact that they did not leave the parental home before reaching 30 years of age. In 2007, the 'university precarious' trajectory grouped together those university graduates with an occupational trajectory characterized by unemployment and precariousness; in 2012, the same type appears, although including a higher proportion of postcompulsory graduates.
Thus, despite the aforementioned methodological precautions, the similarities in the main characteristics of six of the seven types are remarkable. The most significant differences are not to be found, therefore, in the typology composition but in the proportion of young people in each type: broadly speaking, it is important to note that those following a type of trajectory linked to vulnerability and precariousness increased from 25.5 per cent in 2007 12 to 43.6 per cent in 2012. Similarly, the rapidprivileged trajectories ('post-compulsory linear' and 'university linear') declined from 46.1 per cent in 2007 to 27.6 per cent in 2012. Certainly, it could be argued that some of the increased trajectories ('late university' and 'university precarious') may include a diverse set of trajectories. However, the same could be argued with regard to the 'living with parents' trajectories, and linear trajectories also include a diversity of particular pathways. The main issue here is that each of the identified trajectories shows an important internal homogeneity, that is, the main events tend to happen in the same order and moment.
As it may be expected, those young people grouped in the 2012 vulnerable trajectories tend to have a lower social origin, both considering the parents' highest educational level or occupational category. For the same reasons, those young people following more comfortable trajectories tend to have parents with higher degrees and occupational categories. Table A3 in the Appendix shows, for example, that among those young people following a 'working class' trajectory, up to 60.6 per cent have parents who have only reached (if so) compulsory education; the proportion drops to 23.7 per cent among 'university linear' trajectories. Conversely, only 10.2 per cent of the parents of the former work as professionals or technicians, while this proportion reaches 35.5 among the latter.
These results suggest that the economic crisis has not substantially modified the number of typical trajectories but redistributed the young people among them. This stability can be related to two issues: first, one of the new trajectories the crisis may be stimulating is emigration, a type of trajectory that surely can be split into different sub-types (e.g., successful and precarious). However, the Catalan Youth Survey is addressed to Catalan residents and therefore, these emerging trajectories are not included.
Second, among those young people that remain in Catalonia (clearly, the vast majority 13 ), the fact that the number of typical trajectories remains relatively stable after a shift from an economically expansive to a recessive period is probably linked to the impact of the crisis on the extension, reversibility and non-linearity of trajectories. For this reason, it is preferable to analyze this issue in this broader context.

Discussion
This article has offered a case study of the effects of the economic crisis in the trajectories of transition into adulthood. Particularly, the comparison of the trajectories during the economic expansive and recessive periods has shed some light on the nature of the patterns of change in youth transitions that have been found at European level. Thus, it has been shown that the current crisis period has stimulated an 'expansion' of the youth phase, especially with regard to the educational and labour transition. This traditional pattern was characteristic during the expansive period (Serracant, 2011;Van de Velde, 2008) and the current economic crisis has intensified it. However, the extension of transitions is not only the result of young people studying longer or residing longer in the parental home, but it is also the consequence of the rising 'reversibility' of events that the economic crisis has stimulated. At the same time, 'non-linear' trajectories have decreased, which suggests that this change identified at the European level may be related to a greater availability of resources: in an expansive context, young people may tend to make their trajectories more flexible and order their sub-transitions according to their interests, motivations and opportunities, thus altering the traditional pattern of transition. During recession, however, the shared dominant strategy of young people and their families appears to consist of taking each step in due time (when risks are minimized), which also contributes to the extension of trajectories. Finally, a greater 'diversification' has not been found in the types of trajectories young people follow, but there has been a redistribution of the proportion of young people in each group.
Thus, this study case has shown that, in a context of crisis and for the Catalan case, transitions take longer, linear trajectories and reversibility increase and the typical trajectories remain stable. The choice biography hypothesis explains part of the changes in youth trajectories as the outcome of a context of increased opportunities and a reduction of the effect of inequalities linked to the social structure. These transformations would result in an enlarged range of choices available for young people when reflectively constructing their pathways into adulthood. This analysis of the changes in youth trajectories has illustrated, for the Catalan case, that these processes of change are linked to several, often opposed, grounds. The results reveal that economic constraints have increased some of the tendencies (extension and reversibility of trajectories) that, according to the choice biography hypothesis, were linked to a context of improved opportunities. This suggests the need to study, in detail, the nature of these changes in countries with other youth welfare regimes.
Related to this, it is worth highlighting how the economic crisis affects social reproduction in a familistic youth welfare regime. During the economic crisis, the labour market has increased young people's vulnerability and support through public policies has been reduced, especially in those areas directly concerned with youth transitions (most significantly, financial assistance, active employment policies and public housing). In a context where the labour market constitutes a threat and public policies do not represent an opportunity, the family has become even a greater provider of resources, and waiting in the parental home constitutes the dominant strategy among young people in Catalonia. In a typical case of interaction between desires and opportunities (Elster, 2007), young people tend to stay in the parental home to accumulate different types of resources (educational, labour, financial…) to face their transition to adulthood. An increased role of the young people's family was already identified by, for example, by Hutson and Jenkins (1989) when analyzing the United Kingdom's (UK) recession during the 1980s. This greater dependence on the family's material resources and support clearly will reduce equal opportunities in the processes through which young people acquire their social position.