Jump to ContentJump to Main Navigation
Lifelong Learning Policies for Young Adults in EuropeNavigating between Knowledge and Economy$

Marcelo Parreira do Amaral, Siyka Kovacheva, and Xavier Rambla

Print publication date: 2019

Print ISBN-13: 9781447350361

Published to Policy Press Scholarship Online: September 2020

DOI: 10.1332/policypress/9781447350361.001.0001

Show Summary Details
Page of

PRINTED FROM POLICY PRESS SCHOLARSHIP ONLINE (www.policypress.universitypressscholarship.com). (c) Copyright Policy Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in PPSO for personal use.date: 30 July 2021

Assessing young adults’ living conditions across Europe using harmonised quantitative indicators: opportunities and risks for policy makers

Assessing young adults’ living conditions across Europe using harmonised quantitative indicators: opportunities and risks for policy makers

Chapter:
(p.171) 9 Assessing young adults’ living conditions across Europe using harmonised quantitative indicators: opportunities and risks for policy makers
Source:
Lifelong Learning Policies for Young Adults in Europe
Author(s):

Rosario Scandurra

Kristinn Hermannsson

Ruggero Cefalo

, Marcelo Parreira do Amaral, Siyka Kovacheva, Xavier Rambla
Publisher:
Policy Press
DOI:10.1332/policypress/9781447350361.003.0009

Abstract and Keywords

This chapter uses harmonized quantitative regional data on the mediating role of LLL policies in the configuration of individuals living conditions. We focus our attention on four indicators: youth unemployment, tertiary education enrolment, early school leavers and NEET rates. To analyse the determinants of the contextual living conditions we fit persistence models, attempting to explain the status in 2014 with the observed conditions in 2006. We find strong evidence of path dependency. This indicates that the regional contextual living conditions of young adults are overwhelmingly dominated by a combination of the region’s history and developments at the national level. Looking forward, a historically prosperous region in a positive national context is likely to remain so, whilst equally a weak region within a weak national context is likely to remain so. If policy makers are intending to influence the contextual living conditions of young adults, they need to be aware of this inertia. Policies at the national level can be changed and they can be devolved. This could be one way of tackling the inertia, i.e. by providing more policy authority to NUTS-2 regions. Highlighting existing data gaps and improving the availability of territorial information are crucial steps to achieve better targeted policy that isn’t contingent up nation-state-based measures.

Keywords:   Young adults, Living conditions, Quantitative regional data, Living contexts, NUTS-2 level data

Introduction

Understanding the contexts within which young people develop their biographies, the transition to adulthood and the link to the opportunities and constraints structures they provide is becoming increasingly relevant in researching the transition from youth into adulthood. This chapter provides a comparative assessment of key measures of young adults’ contextual living conditions in European regions. This is part of research carried out within YOUNG_ ADULLLT. The main aim of the chapter is to describe a framework of analysis for assessing young adults’ living conditions, demonstrate the pros and cons of such an approach and to report some key results about young adults’ living conditions.1 We derive our results from a selection within a wide range of socioeconomic indicators on the specific living conditions of young adults, focusing on lifelong learning (LLL) and inclusion in education and the labour market as these data are widely used to steer policy-making. Specifically, we aim to disentangle the territorial dimension of contextual living conditions in which young adults are inserted in relation to LLL, in order to understand how regional contexts interact with dynamics of growth and social inclusion. In this sense, living conditions are understood in a comprehensive manner as a plurality of aspects within which individual life courses unfold, such as socioeconomic conditions, the labour market, education and training systems and well-being.

This does not imply a deterministic view where the context and structural factors completely prevail over individual agency and (p.172) self-determination. However, it stresses the relevance of contextual living conditions in building different structures of opportunities for young people, in terms of a complex mix of enabling and constraining features, according to the place where they live. The results contribute to building the contextual structure of enablements and constraints with which young people engage to actively form their dispositions and choices.

This contribution also represents an effort to move away from the nation state as a basic unit of analysis. Global evaluation assessments are reinforcing methodological nationalism by considering mean average of performance. This is particularly the case for educational studies (e.g. PISA (Programme for International Student Assessment) or TIMSS (Trends in International Mathematics and Science Study)). Our approach is in response to a general tendency in the social sciences, and in comparative education in particular, to adopt the nation state as an assumed unit of analysis (Wimmer and Glick Schiller, 2003; Robertson, 2011). This is related to different factors such as the construction of the nation state, its historical tradition and the establishment of statistics as a discipline (Porter, 1996; Desrosières, 2008). Furthermore, we observe that the increasing influence of transnational organisations and the availability of international assessments have altered the way governments and education stakeholders think, discuss and make decisions about education systems (Thévenot, 2009; Mundy et al, 2016; Normand, 2016). The influence of these assessments has not escaped criticism from the comparative education community, although this criticism has revolved predominantly around the political use and interpretation of this data (Grek, 2016). This chapter presents some of key measures of young adults’ living conditions in Europe, providing a country-average picture. This is very much connected to a branch of literature on regional cohesion policy which has concentrated mainly on economic growth in European Union (EU) territories or which has traditionally focused on different area of policy (e.g. agriculture). However, efforts to produce context-based measures for assessing living conditions at a regional level are still fragmented, reflecting the neglected importance of territorial differences.

Constructing a framework of analysis for assessing young adults’ living conditions

In contemporary societies, young individuals face uncertainty in the transition to adulthood and labour market entry, as well as in the phase of family formation, leading them to be labelled the ‘losers’ of (p.173) globalization processes (Buchholz et al, 2009). The result is a life course often characterized by uncertain access to material resources and the fragility of family and social networks (Blossfeld and Hofäcker, 2014). There is broad consensus that young people’s transitions from youth to adulthood are undergoing de-standardization, individualization and fragmentation (Biggart and Walther, 2006).

Difficulties experienced in the transition from school to work are usually deemed as particularly relevant in this regard. By the end of the 1970s, the match between labour demand and supply had become more problematic as macroeconomic policy in Europe shifted away from a full employment imperative to a low inflation imperative. Many factors conspire in the difficulties experienced by young adults in accessing employment. First, the ongoing flexibilisation of the labour market brings about the spread of temporary and non-standard work arrangements (as opposed to a standard working relationship based on a full-time and permanent contract). This has increased the risk of being trapped in low-income and precarious dead-end jobs, with negative long-term effects on individual working biographies and future pensions (Cuzzocrea, 2014). Second, the trend of tertiarisation and the expansion of high productivity economic sectors imply a stronger disadvantage for people possessing low or obsolete skills, who mostly end up as unemployed or employed in the low value-added service sector, depicting a typical post-industrial employment problem (Bonoli, 2012). One could object that younger generations are on average better educated than older cohorts. However, and here we come to the third factor implied, when caught in the school-to-work transition phase they often lack job experience requested by employers, nor do they possess strong ties with social partners and consequently strategic power for negotiation. Therefore, stable employment in permanent and well-paid jobs is quite hard to achieve for young labour market entrants. Ryan (2008) refers to this paradoxical disadvantage as a double skill bias, as it refers both to low skills and to the lack of job-related and soft skills that can be fully developed through work experience. In the literature on labour market participation and growing inequalities, young people are often considered as outsiders, a group characterized by disadvantaged conditions and less opportunities with respect to other groups of insiders such as, for instance, middle-aged males with a permanent working position (Lindbeck and Snower, 2001; Emmenegger et al, 2012). This condition is exacerbated by demographic changes that weaken the caring capacities of families (population ageing, low fertility rates and diffusion of new family models), as well as by the slow adaptation of welfare programmes to the (p.174) changing configuration of risk profiles (Ferrera, 1996; Bonoli, 2005). However, such a general trend is mediated by varying configurations of the interface among the education system, the labour market and the welfare state that influences young individuals’ opportunities and constraints, as debated in the literature on LLL (Rubenson, 2006; Rubenson and Desjardins, 2009; Blossfeld et al, 2014; Lehmann, 2014). In this light, Verdier (2012) builds a typology of public policies’ regimes of LLL, stressing the relevance of each national context, while Pastore (2011) draws upon the literature on comparative welfare states describing related ‘worlds’ of school-to-work transitions. In a similar fashion, Walther (2006) looks at different transition regimes, identifying variations in the interplay between specific contextual structures and agency expressed by young people’s subjective perspectives. The relationship between structural reproduction and the actual decision-making of individuals was highly debated in stratification research and youth transition research, with the latter criticizing the overemphasis and relevance of the capacity of institutional structure to reinforce inequality and produce vulnerability, thus stressing the concept of agency (Lehmann, 2014). Without neglecting the influence of social structures, scholars state that personal agency is always present in the transition from youth to adulthood: young people can actively shape some important dimensions of their experience, as they make distinctive choices about their education and career pathways at critical junctures (Anisef et al, 2000).

In our understanding, living conditions refers to the exposure to social disadvantage coming from complex configurations of risks affecting various life domains. It is a ‘fluctuating’ condition of weak social integration and high insecurity (Castel, 2000) that overlaps only partially with the identification of socially excluded groups characterized by material deprivation. In this light, scholars investigating economic insecurity (which is a component of social insecurity and vulnerability) argue that the growing inequality in present European societies poses threats not only at the bottom of the income distribution, but also in the traditionally protected and secure middle classes (Mau et al, 2012; Ranci et al, 2014). Therefore, we underline the necessity of investigating people’s material living conditions (OECD, 2017) in their specific contexts (Pawson and Tilley, 1997; Kazepov and Ranci, 2016), as these are strongly connected to their degree of social integration. Institutional structures and specific local characteristics are important mediators in shaping young people’s lives. In this chapter, we investigate the contextual living conditions through a set of key indicators that shape young adults’ (p.175) opportunity structures (Kerckhoff, 1995, 2001; Cloward and Ohlin, 2013; Lehmann, 2014) in the regions where they live and build their life trajectories. These are the tertiary education attainment (TEA) rate, the early school leavers (ESL) rate, the youth unemployment rate (YUR) and the not in employment, education or training (NEET) rate.

Comparative analyses of inequality, poverty and vulnerability have mainly taken individuals or countries as their unit of analysis (Ranci, 2010) while less attention has been devoted to contextual and place-based approaches. However, several recent phenomena have directed attention towards regional and local levels of analysis: processes of European integration and rescaling limited the role of the central state and at the same time attributed greater relevance to subnational scales of governance (Kazepov, 2010), while marked and persisting regional and territorial disparities emerged within European countries, as the multifaceted debate on territorial cohesion demonstrates (Medeiros, 2016). In this light, Atkinson and colleagues (2002) stress the importance of regional and place-based indicators, particularly when considering a wider view of exclusion that covers more dimensions, including poverty, education and health. This implies taking into consideration the interplay among contextual factors as a manifestation of socioeconomic trends in the region and the impact of institutional factors related to welfare provision and structures of multi-level governance. Therefore, we focus on the contextual living conditions in selected regions (Nomenclature of Territorial Units of Statistics (NUTS) 2 level), and their evolution and the variation within each EU country. What we want to stress is the relevance of contextual living conditions in building different structures of opportunities for young people, in terms of complex mixes of enablements and constraints, according to the place where they live. Our chapter contributes to building a picture of the contextual structure of enablements and constraints with which young people engage to actively form their dispositions and choices.

The regional level as an appropriate unit of analysis

To gain understanding of the heterogeneity of young adults’ living conditions, we use the regional level as basic unit of analysis (see Chapter 2, in this volume). Within the EU, the official statistical approach to gathering data on structural information uses a hierarchical categorisation of EU territories and regions. As a geographical system, a division was developed by Eurostat to structure and classify regional (p.176) statistics resulting in NUTS. The aim is to provide a single and coherent system for ‘comparable and harmonised data for the European Union to use in the definition, implementation and analysis of Community policies’ (Eurostat, 2007: 3). This is relevant as, due to changing realities such as internationalisation, Europeanisation and globalisation, the concept of using administrative units, in particular those at the national level, as a unit of analysis is increasingly questioned as a useful tool to describe social realities (Weiler et al, 2017: 10). The YOUNG_ ADULLLT project builds on the perspective that the implementation of LLL policies is best studied at the regional/local level to understand the context specificity of young adult life courses beyond the national level. Therefore, the use of the concept of functional regions (FR) sharpens the focus on regional differences and variations. However, using the concept also raises challenges for the validity of research, as the different FRs can be a (mis)match with the territorial and/or administrative regions that are predominantly used within established statistics, as well as creating challenges in data availability of different sources. For example, statistical data on socioeconomic and sociodemographic aspects, education and training, labour market and welfare dimensions are not limited to administrative units (countries, states, districts, provinces or cities). Departing from the tension among official descriptions of communities, changing realities and data availability, we deal with this in two ways: first, by developing a practical approach of data collection, and, second, by assessing the data production process of the EU. In the case of the latter, the data gaps in the European Statistical System also imply how data are collected within the EU with regard to our FRs. This provides insight into the question of how data are used to steer political processes related to LLL policies and thus the process of definition, coordination and implementation of policy measures. In the case of the data-gathering process, the data collected are the closest possible to the regional level. In this way, pre-existing data on NUTS 2 was used, albeit enriched and specified by local/regional sources. This is relevant as subdivisions in some levels do not necessarily correspond to administrative divisions within the country. The level of analysis is constrained by the existing territorial division, which reflects the data availability.2

Considering this mismatch between the territories selected in the project as FRs and the availability of the data extracted from international data sources, the level of analysis varies hugely in terms of percentage of young adults living in these regions. The units of analysis (NUTS 2 regions) vary in terms of territorial extension and rural versus urban as displayed by their degree of urbanization. In the (p.177) project sample, 18 regions were selected. Vienna and Bremen are both highly populated and dense areas and correspond administratively to single federal states (Bundesländer). Other regions such as Andalusia or Pohjois-ja Itä-Suomi represent, respectively, 17 and 67 per cent of the entire territory of Spain and Finland. Other regions, such as Alentejo or both the Finnish regions selected, are large and rural. This substantially influences the estimates of the overall findings and needs to be taken into consideration when interpreting the results. It is natural to expect that urban areas are richer in terms of labour market opportunities or show a higher degree of economic innovation. Moreover, there have been big changes in terms of the share of young people living in these regions as a result of the different economic circumstances these territories experienced during 2007–14. South European regions such as Spain, Portugal, Italy, Croatia and also Bulgaria suffered a loss in the share of the young adult population, ranging from –39 per cent in Catalonia to –2.7 per cent in Yugozapaden. Similar or even more extreme results are shown if we consider the population between 20 and 34 years old. This is partially related to ageing processes, although migration flows are also an influence in some cases.

Data collation and the operationalisation process

This subsection describes the methodology adopted and the operationalisation process carried out when conducting quantitative research on young adults’ contextual living conditions. First, the researchers designed a framework of analysis and selected the dimensions and categories of interest for the overall research. Next, the indicators connected to the categories were selected (see Figure 9.1 and Table 9.A2 for a detailed description of the items).

Third, administrative sources and comparative surveys were identified and the data coverage and quality at national and regional levels assessed. After considering data availability constraints, the level of analysis selected was NUTS 2, which represents the highest level of territorial disaggregation to conduct an in-depth analysis of young adults’ living conditions.

The data collation draws on databases from national administrative sources and comparative surveys compiled by international organisations such as Eurostat and the Organisation for Economic Co-operation and Development – the main sources of the European Labour Force Survey (LFS) and the European Union Statistics on Income and Living Conditions (EU-SILC). In Figure 9.1, the six dimensions of the analysis are show with the correspondent item components. (p.178)

Assessing young adults’ living conditions across Europe using harmonised quantitative indicators: opportunities and risks for policy makers

Figure 9.1: Dimensions and indicators selected

Data was collated for a span of more than ten years, from 2005 through 2016, the latest available year. This enables comparability across countries and regions, before and after the Great Recession. Young adults are defined as individuals aged between 18 and 29 years, however, a plurality of age ranges were used pragmatically to overcome data limitations and select sound indicators. For example, for the category of ‘attainment’ we select the age group 30 and 34 years. This choice was made because the length of education programmes vary largely across the EU and it is better to peak the age group at the point where most of the population has finished their post-compulsory and post-tertiary education experiences.

Limitations and constraints of the analysis

The limitations of the research are diverse and need to be considered carefully. Like all concepts in the social sciences, and academic disciplines in general, the act of constructing measurements implies a selection of the dimensions (in Ancient Greek κατηγορια‎ or in Latin categoria) which should be operationalised and thus leads to a simplification of the object of study. This means a transformation of some qualities into metrics, which is not just a technical process, but an important feature of social life (Hacking, 1999; Desrosières, 2008). This process is generally called commensuration and has been widely (p.179) examined by historians, statisticians, sociologists and philosophers (Espeland and Stevens, 1998). From Plato and Aristotle, to Marx, Weber, Simmel and Foucault, the implications of commensuration have been analysed as a process that influences our valuation and the way we invest in goods and services. In the field of education as an example, questions of performance and its evaluation have gained greater social and scholarly prominence in recent years. With the spread of market fundamentalism and the new public management turn, governments have created tools to ensure greater efficacy, with the result that quantitative measures of performance and benchmarking are spreading and are having important configuring effects on a range of institutions and domains of human activity (Lamont 2012; Didier, 2007).

Undoubtedly, our research could not escape the process of commensuration. First, the establishment, recognition and use of a statistical object is very appealing. Second, the interpretation and political use of each measure is a very powerful way to push forward a specific approach or even a political agenda (Meyer and Benavot, 2013). In this sense, our research objectives are constrained by existing and available sources and their comparability, and by statistical issues such as representativeness. On this matter, we stress that the most relevant survey data source for the research objectives is the European LFS, as the only survey available and comparable at NUTS 2 level that collects information on the living conditions of young adults.

There are substantial limitations in the availability of complete information on young adults’ living conditions at regional and subnational levels. The Eurostat statistical information system relies on restricted administrative records with territorial disaggregation, mainly on economics, demographics and the health system. Moreover, this information is quite dispersed and not very user-friendly. Few micro-data sources provide a scattered figure on territorial differences in young adults’ living conditions. The most complete information available for this purpose is the LFS.

An important limitation to producing regional indicators on young adults’ living conditions is the absence of complete information on the sample structure and territorial identification both in the EU-SILC and LFS, which are potentially the most adequate data sources for the research objective. This limited the ability of the YOUNG_ADULLLT project to derive local-level indicators from these sources. FRs partially correspond with the NUTS 2 classification and some indicators are available at this level. However, deriving finer contextual-based measures of young adults and LLL policies in the European territories (p.180) is particularly challenging, as few data are available at the NUTS 3 level.

Findings

The findings are organized as follows: first, we report the distribution of four key indicators of young adults’ living conditions (Figures 9.2, 9.3, 9.4 and 9.5): TEA, ESL, YUR and NEET. As a second step, we show the evolution of these indicators after 2006 (Table 9.2). As a third step, we report the persistence (or elasticity) over the selected period, which shows how the level of each single indicator at the beginning of the period explains the level in the last period (Table 9.3). Thus, we explore the path dependency of each indicator over the period.

Figure 9.2 plots the country average and the range within each country’s regions of the share of people aged 30–34 years with tertiary education attainment (International Standard Classification of Education [ISCED] 5–8). The extremes of the bars represent each country’s regions lowest and highest share of tertiary-educated people. The following figures represent the same estimates as previously described and the summary statistics are given in Table 9.A1. Countries represented with a dot are the smallest countries, which are composed of a single territorial unit,3 for example Malta, Cyprus, Slovenia or Luxembourg. In Lithuania, Luxembourg, Cyprus, Ireland and Sweden more than one out of two people aged between 30 and 34 years attained tertiary education, while in Italy and Romania only one out of four attained a tertiary degree. However, as the figure shows there is a high variation in the share of tertiary-educated young adults within each country, with the highest coefficient of variation in Slovakia, Romania, Czech Republic, Bulgaria, Greece, Hungary and Denmark.4 In these countries the share of tertiary-educated young adults varies more compared to the rest of the European countries.

Figure 9.3 plots the country average and the range within each country’s regions of the share of ESL. In Spain and Malta more than one out of five people aged between 18 and 24 years is an ESL, while in Croatia, Slovenia, Poland and the Czech Republic not quite 5 per cent or less of these populations is an ESL. However, the country average masks a high regional variation in ESL rates, with the highest variability found in the Czech Republic, Poland, Greece and Bulgaria. In these countries the share of tertiary-educated young adults varies more compared to the rest of the European countries, while it is much reduced in Sweden and Denmark.

(p.181)

Assessing young adults’ living conditions across Europe using harmonised quantitative indicators: opportunities and risks for policy makers

Figure 9.2: TEA, ISCED 5–8, % population, aged 30–34

Note: AT: Austria, BE: Belgium, BG: Bulgaria, CY: Cyprus, CZ: Czech Republic, DE: Germany, DK: Denmark, EE: Estonia, EL: Greece, ES: Spain, FI: Finland, FR: France, HR: Croatia, HU: Hungary, IE: Ireland, IT: Italy, LT: Lithuania, LU: Luxembourg, LV: Latvia, MT: Malta, NL: Netherlands, PL: Poland, PT: Portugal, RO: Romania, SE: Sweden, SL: Slovenia, SK: Slovakia, UK: United Kingdom.

(p.182)

Assessing young adults’ living conditions across Europe using harmonised quantitative indicators: opportunities and risks for policy makers

Figure 9.3: ESL, % population, aged 18–24

(p.183) Figure 9.4 plots the country average and the range within each country’s regions of the share of YUR. On average, YUR affects more than 42 per cent of the youth population in Spain, Greece, Croatia and Italy, while it is comparatively very low in Germany, Austria, Malta, Denmark and the Netherlands. The countries with the highest variability in YUR are France, Austria, Belgium and Denmark, while Croatia, Denmark, Sweden, Finland and Spain report very low regional disparities. Among the countries with the highest share of YUR, Greece and Italy are the ones with the highest regional differences, both reflecting the territorial differentiation that pre-existed the Great Recession and has since widened.

Figure 9.5 shows the NEET rate of the 18–24-year-old population. This indicator is closely connected to YUR, although the age interval differs slightly.

NEET rate is a broad indicator compared to YUR, as suggested by Furlong (2017), grouping together those who are inactive, single mothers or disabled. In Italy, Greece, Croatia, Cyprus and Bulgaria more than one out of four young adults are NEET. In the Netherlands, Denmark, Luxembourg, Germany, Sweden and Austria less than one out of ten young adults are NEET. The biggest difference between regions in NEET share is found in Italy, France and Bulgaria; however, the highest overall variability is found in the former two countries and Portugal.

As a second step of the analysis we explore the evolution of these indicators since 2006. Table 9.1 shows the estimates of a simple version of the model, examining the trend of four key indicators of young adults’ living conditions.

We observe that, on average over the period, YUR was 20 per cent. This decreased in 2007 and 2008 by 3 per cent. After 2009 YUR increased steadily, reaching a peak in 2013. The trend of this indicator is one of the most visible consequences of the Great Recession in Europe, which has pushed a great number of young adults out of the EU labour market. During the same period education attainment increased, such that the share of young adults who hold a tertiary education degree was 26.8 per cent, more than a quarter of the entire population. The progress registered during this period is very relevant, with an almost linear increase which reached 9.2 per cent in 2014. When examining ESL and NEET rates over the period, we found both indicators were at a similar level (15 per cent), although the pattern of the evolution is very different. ESL seems to have decreased linearly from 2009 until 2014, while NEET has followed a pattern similar to YUR – although the size of the coefficients differs – decreasing between 2006 and 2008 and then increasing after the Great Recession.

(p.184)

Assessing young adults’ living conditions across Europe using harmonised quantitative indicators: opportunities and risks for policy makers

Figure 9.4: YUR, % population, aged 15–24

(p.185)

Assessing young adults’ living conditions across Europe using harmonised quantitative indicators: opportunities and risks for policy makers

Figure 9.5: NEET, % population, aged 18–24

(p.186)

Table 9.1: Annual average changes for NUTS 2 regions since 2006

Variables

(1)

(2)

(3)

(4)

YUR, 15–24 years

TEA, 30–34 years

ESL, 18–24 years

NEET, 18–24 years

2006

–1.271***

1.084***

–0.206

–1.215***

(0.226)

(0.175)

(0.135)

(0.140)

2007

–3.350***

1.996***

–0.00221

–1.702***

(0.318)

(0.239)

(0.220)

(0.235)

2008

–3.288***

3.298***

–0.341

–1.842***

(0.412)

(0.244)

(0.228)

(0.289)

2009

0.851

4.566***

–0.969***

0.125

(0.520)

(0.287)

(0.244)

(0.322)

2010

2.604***

5.422***

–1.215***

0.700**

(0.553)

(0.279)

(0.241)

(0.323)

2011

3.861***

6.298***

–1.750***

1.053***

(0.639)

(0.320)

(0.286)

(0.356)

2012

6.072***

7.466***

–2.472***

1.603***

(0.752)

(0.348)

(0.296)

(0.369)

2013

6.906***

8.193***

–3.210***

1.548***

(0.807)

(0.366)

(0.304)

(0.382)

2014

5.227***

9.231***

–3.979***

0.810**

(0.797)

(0.415)

(0.324)

(0.365)

2015

3.785***

0.303

(0.788)

(0.363)

Constant

19.99***

26.87***

15.18***

15.75***

(0.622)

(0.634)

(0.538)

(0.428)

Observations

2,770

2,552

2,496

2,853

Number of NUTS 2

271

261

261

272

Note: Robust standard errors in parentheses; (***) p<0.01,

(**) p<0.05,

* p<0.1.

Table 9.2 presents estimates of a simple version of this model, including and omitting country fixed effects. In our specifications we include the persistence term for every indicator at the beginning of the period (e.g. 2006). All the models show strong persistence effects. Interpreted as a predictive model, it suggests that in the absence of any other influences the YUR in 2014 will amount to approximately 82 per cent of the level of YUR in 2006. However, the R2 varies depending on the indicator. Persistence effects only explain 39 per cent of the variation in youth unemployment. However, these can explain more than two thirds of the variation in NEET rates and educational attainment.

(p.187)

Table 9.2: Persistence (or elasticity) over the selected period

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

YUR

YUR

TEA

TEA

ESL

ESL

NEET

NEET

YUR, 2006

0.823***

0.656***

(0.0577)

(0.0456)

TEA, 2006

0.573***

0.716***

(0.0251)

(0.0362)

ESL, 2006

0.709***

0.765***

(0.0257)

(0.0461)

NEET, 2006

0.765***

0.666***

(0.0424)

(0.0424)

Constant

0.732***

0.813***

1.684***

1.486***

0.504***

0.148

0.714***

0.719***

(0.174)

(0.120)

(0.0850)

(0.118)

(0.0700)

(0.129)

(0.114)

(0.0998)

Observations

271

271

261

261

261

261

272

272

R2

0.391

0.925

0.653

0.838

0.677

0.801

0.452

0.881

Country FE

No

Yes

No

Yes

No

Yes

No

Yes

R2 Adj

0.390

0.924

0.653

0.837

0.677

0.799

0.452

0.879

Note: Robust standard errors in parentheses; (***) p<0.01,

** p<0.05,

* p<0.1.FE:fixed effects;

R2: coefficient of determination.

(p.188) In the second specification of this model we control for country fixed effects, which soak out the unobserved heterogeneity, removing the first difference among the countries considered. The persistence is reduced to 65 per cent. However, the overall variance explained increased and now approaches 1, which demonstrates that the confluence of the regional past and the country-level present dominate the development of young adults’ contextual living conditions. The sizes of the persistence effects differ across the indicators selected; however, these are very high: 71.6 per cent for TEA, 76.5 per cent for ESL and 66.6 per cent for NEET, when country fixed effects are included. These show how strongly the initial level of such indicators impacted the 2014 level, showing very high path dependency in young adults across European regions.

Conclusion

This chapter emphasizes the relevance of contextual living conditions in shaping the structures of opportunities for young adults in different regional settings. It provides synthetic information on different dimensions that can be usefully related to LLL policy-making and to the impact of such interventions. Given its broad range, the secondary data analysis presented has to be seen as a contribution to a wider strategy integrating quantitative results as a basis for the institutional and policy analysis carried out in YOUNG_ADULLLT (Scandurra et al, 2018).

The research uses harmonized quantitative data on the mediating role of LLL policies in the configuration of individuals’ living conditions, getting as close as possible to the regional level using pre-existing data sets. Furthermore, it explores data gaps in the European Statistical System to complement those data with context-specific information. The findings show that there are huge differences both in the level and dispersal of young adults’ living conditions across European territories. However, this evidence is partial and relies on limited and aggregate information.

We focus our attention on four indicators: youth unemployment, tertiary education enrolment, ESL and NEET rates. To analyse the determinants of contextual living conditions we designed simple persistence models, attempting to explain the status in 2014 using the observed conditions in 2006. We find strong evidence of path dependency and once we introduce national-level fixed effects the models show an extremely high R2. Taken together, this indicates that the regional contextual living conditions of young adults are (p.189) overwhelmingly dominated by a combination of the region’s history and developments at the national level. Looking ahead, a historically prosperous region in a positive national context is likely to retain this status, whilst equally, a weak region within a weak national context is likely to remain weak. If policy makers are intending to influence the contextual living conditions of young adults, they need to be aware of this inertia. Policies at the national level can be changed and can be devolved. This could be one way of tackling the inertia, that is, by providing more policy authority to NUTS 2 regions. However, history cannot be changed and therefore policy makers need to take this into account when formulating expectations as to how much transformation can reasonably be expected.

In order to better inform policies, an intense effort is needed to develop richer context-based information at a territorial level below NUTS 2. Highlighting existing data gaps and improving the availability of territorial information are crucial steps to achieving better targeted policy that is not contingent on nation-state-based measures. Due to changing realities, such as internationalisation, Europeanisation and globalisation processes, the use of the national level as a representative unit of account should be questioned, and more localised measures could be useful tools to describe changing social realities.

There is a need to increase social impact by understanding the role of the specific contexts within which measures are implemented. This calls for more contextualized information, which is a prerequisite for regional comparative analysis and more targeted and evidence-based policy. Moreover, to develop a broader interpretative framework, it is necessary to tap new data sources that are not strictly based on existing measures of education and labour market status. A holistic approach to living conditions is needed, particularly in a time of socioeconomic changes and reconfiguration of young adults’ motivations and aspirations.

References

Bibliography references:

Anisef, P., Axelrod, P., Baichman-Anisef, E., James, C. and Turrittin, A. (2000) Opportunity and Uncertainty: Life Course Experiences of the Class of ’73, Toronto: University of Toronto Press.

Atkinson, A., Cantillon, B., Marlier, E. and Nolan, B. (2002) Social Indicators: The EU and Social Inclusion, Oxford: Oxford University Press.

Biggart, A. and Walther, A. (2006) ‘Coping with yo-yo transitions: Young adults’ struggle for support – between family and state in comparative perspective’, in C. Leccardi and E. Ruspini (eds) A New Youth? Young People, Generations and Family Life, Aldershot: Ashgate, pp 41–62.

Blossfeld, H.-P. and Hofäcker, D. (2014) ‘Globalization, rising uncertainty and life courses in modern societies: A summary of research findings and open research questions’, Sociologia del lavoro, 136: 16–33.

Blossfeld, H.-P., Kilpi-Jakonen, E., Vono De Vilhena, D. and Buchholz, S. (eds) (2014) Adult Learning in Modern Societies: An International Comparison from a Life-Course Perspective, Cheltenham: Edward Elgar Publishing.

Bonoli, G. (2005) ‘The politics of the new social policies: Providing coverage against new social risks in mature welfare states’, Policy & Politics, 33(3): 431–49.

Bonoli, G. (2012) ‘The postindustrial employment problem and active labour market policy’, 10th ESPAnet Annual Conference, Edinburgh.

Buchholz, S., Hofäcker, D., Mills, M., Blossfeld, H.P., Kurz, K. and Hofmeister, H. (2009) ‘Life courses in the globalization process: The development of social inequalities in modern societies’, European Sociological Review, 25(1): 53–71.

Castel, R. (2000) ‘The roads to disaffiliation: Insecure work and vulnerable relationships’, International Journal of Urban and Regional Research, 24(3): 519–35.

Cloward, R. A. and Ohlin, L. E. (2013). Delinquency and Opportunity: A Study of Delinquent Gangs (vol. 6), London: Routledge.

(p.191) Cuzzocrea, V. (2014). ‘Projecting the category of NEET into the future’, in Council of Europe (ed) Perspectives on Youth: Volume 1 2020 – What Do YOU, Strasbourg: Council of Europe, pp 69–82.

Didier, E. (2007) ‘Do statistics “perform” the economy?’, in D. MacKenzie, F. Muniesa and L. Siu (eds) Do Economists Make Markets? On the Performativity of Economics, Princeton, NJ: Princeton University Press, pp 276-310.

Desrosières, A. 2008) ‘La statistique, outil de gouvernement et outil de prevue’, in A. Desrosières (ed), L’argument Statistique I: Pour Une Sociologie Historique de La Quantification, Paris: Presse s des Mines, pp 7–20.

Emmenegger, P., Häusermann, S., Palier, B. and Seeleib-Kaiser, M. (eds) (2012) The Age of Dualization: The Changing Face of Inequality in Deindustrializing Societies, Oxford: Oxford University Press.

Espeland, W. N. and Stevens, M. L. (1998) ‘Commensuration’, Annual Review of Sociology, 24: 313–43.

Eurostat (2007) ‘Regions in the European Union Nomenclature of Territorial Units for Statistics NUTS2006 /EU-27’, Eurostat. Retrieved at: https://ec.europa.eu/eurostat/web/products-manuals-and-guidelines/-/KS-RA-07-020

Ferrera, M. (1996) ‘The southern model of welfare in social Europe’, Journal of European Social Policy, 6(1): 17–37.

Furlong, A. (ed) (2017) Routledge Handbook of Youth and Young Adulthood, London: Taylor & Francis.

Grek, S. (2016). ‘Knowledge actors and the construction of new governing panoramas: The case of the European Commission’s DG education and culture’ [Acteurs de connaissances et la construction des nouveaux panoramas du gouvernement], Le cas de la Direction Générale de l’Édu. Educacao E Sociedade, 37(136), available from: https://doi.org/10.1590/es0101-73302016166099

Hacking, I. (1999) The Social Construction of What? Cambridge, MA: Harvard University Press.

Kazepov, Y. (ed) (2010) Rescaling Social Policies: Towards Multilevel Governance in Europe, Aldershot: Ashgate Publishing.

Kazepov, Y. and Ranci, C. (2016) ‘Making social investment work: A comparative analysis of the socio-economic preconditions of social investment policies’, Annual ESPANnet 2016 Conference, Rotterdam.

Kerckhoff, A. C. (1995) ‘Institutional arrangements and stratification processes in industrial societies’, Annual Review of Sociology, 21(1): 323–47.

(p.192) Kerckhoff, A. C. (2001) ‘Education and social stratification processes in comparative perspective’, Sociology of Education, 74: 3–18.

Lamont (2012) ‘Toward a Comparative sociology of valuation and evaluation’, Annual Review of Sociology, 38: 201–21.

Lehmann, W. (2014). Choosing to Labour? School–Work Transitions and Social Class, Montreal: McGill-Queen’s University Press.

Lindbeck, A. and Snower, D. J. (2001) ‘Insiders versus outsiders’, Journal of Economic Perspectives, 15(1): 165–88.

Mau, S., Mewes, J. and Schöneck, N. M. (2012) ‘What determines subjective socioeconomic insecurity? Context and class in comparative perspective’, Socio-Economic Review, 10(4): 655–82.

Medeiros, E. (2016) ‘Territorial cohesion: An EU concept’, European Journal of Spatial Development, 60: 1–30.

Meyer, H. D. and Benavot, A. (2013) PISA, Power, and Policy: The Emergence of Global Educational Governance. Oxford: Symposium Books.

Mundy, K., Green, A., Lingard, B. and Verger, A. (2016) ‘Introduction: The globalization of education policy – key approaches and debates’, in K. Mundy, A. Green, B. Lingard and A. Verger (eds) The Handbook of Global Education Policy, Hoboken, NJ: John Wiley & Sons, pp 1–20.

Normand, R. (2016) The Changing Epistemic Governance of European Education: The Fabrication of the Homo Academicus Europeanus? Cham: Springer International Publishing.

OECD (Organisation for Economic Co-operation and Development) (2017) How is Life? 2017: Measuring Well-Being, Paris: OECD.

Pastore, F. (2011) Fuori dal tunnel. Le difficili transizioni dalla scuola al lavoro in Italia e nel mondo, Turin: Giappichelli.

Pawson, R. and Tilley, N. (1997) Realistic Evaluation, London: SAGE.

Porter, T. M. (1996) Trust in Numbers: The Pursuit of Objectivity in Science and Public Life, New York and London: Princeton University Press.

Ranci, C. (2010). ‘Bringing territory back in social comparative research’, in C. Ranci (ed) Social Vulnerability in Europe: The New Configuration of Social Risks, Basingstoke and New York: Palgrave Macmillan, pp 25–34.

Ranci, C., Brandsen, T. and Sabatinelli, S. (2014) Social Vulnerability in European Cities: The Role of Local Welfare in Times of Crisis, Basingstoke: Palgrave Macmillan.

Robertson, S. L. (2011) ‘The new spatial politics of (re) bordering and (re) ordering the state–education–citizen relation’, International Review of Education, 57 (3–4): 277–97.

(p.193) Rubenson, K. (2006) ‘The Nordic model of lifelong learning’, Compare: A Journal of Comparative and International Education, 36(3): 327–41.

Rubenson, K. and Desjardins, R. (2009) ‘The impact of welfare state regimes on barriers to participation in adult education’, Adult Education Quarterly, 59(3): 187–207.

Ryan, P. (2008) ‘Youth employment problems and school-to-work institutions in advanced economies’, in G. De Freitas (ed), Young Workers in the Global Economy: Job Challenges in North America, Europe and Japan, Cheltenham: Edward Elgar, pp 137–59.

Scandurra, R., Cefalo, R., Hermannson, K. and Kazepov, Y. (2018) ‘Cross-national and cross-regional report quantitative analysis’, available from: www.young-adulllt.eu/publications/working-paper/YOUNG_ADULLLT_Deliverable_D4_2_Crossnational-and-crossregional-Report-Quantitative-Analysis.pdf?m=1530517072

Thévenot, L. (2009) ‘Governing life by standards: A view from engagements’, Social Studies of Science, 39(5): 793–813.

Verdier, É. (2012) ‘European lifelong learning strategy and diversity of national devices: An interpretation in terms of public policy regimes’, Scuola Democratica, 4: 1–20.

Walther, A. (2006) ‘Regimes of youth transitions: Choice, flexibility and security in young people’s experiences across different European contexts’, Young, 14(2): 119–39.

Weiler, A., Schaufler, S., Parreira do Amaral, M., Ganter de Otero, J. P., Wutzkowsky, F., Valiente, O., Neves, T. and Ribeiro, A. B. (2017) ‘Launching and research design: State of the art report’, YOUNG_ADULLLT Working Paper, University of Münster.

Wimmer, A. and Glick Schiller, N. (2003) ‘Methodological nationalism, the social sciences, and the study of migration: An essay in historical epistemology’, International Migration Review, 37(3): 576–610.

See Tables 9.A1 and 9.A2.

(p.194)

(p.195)

Table 9.A1: Summary statistics of selected indicators

County

ESL, 18–24 years

NEET, 18–24 years

Youth unemployment, 15–24 years

Ed. attainment 30–34 ISCED 5–8

Mean

CV

MAX

MIN

SD

Mean

CV

MAX

MIN

SD

Mean

CV

MAX

MIN

SD

Mean

CV

MAX

MIN

SD

AT

7.00

0.23

9.00

4.70

1.63

9.40

0.19

12.40

7.10

1.75

10.30

0.41

18.00

6.00

4.25

40.00

0.16

50.40

31.40

6.21

BE

9.80

0.36

15.20

5.40

3.53

15.00

0.26

21.70

9.70

3.95

23.20

0.40

39.50

13.20

9.17

43.80

0.15

55.50

34.00

6.39

BG

12.90

0.40

20.80

5.90

5.15

24.50

0.45

45.70

12.40

10.96

23.80

0.20

28.90

16.80

4.66

30.90

0.27

43.00

21.70

8.40

CY

6.80

6.80

6.80

25.10

25.10

25.10

36.00

36.00

36.00

52.50

52.50

52.50

CZ

5.50

0.58

12.80

2.50

3.19

10.50

0.33

17.30

5.60

3.45

15.90

0.24

22.90

10.10

3.84

28.20

0.31

45.00

16.00

8.72

DE

9.50

0.26

14.00

5.40

2.47

8.90

0.26

13.60

5.10

2.31

7.70

0.35

15.50

3.70

2.69

31.40

0.21

45.50

20.40

6.64

DK

7.80

0.10

9.00

7.00

0.76

7.80

0.13

8.90

6.60

1.00

12.60

0.08

14.30

11.50

1.00

44.90

0.25

59.30

31.10

11.10

EE

11.40

11.40

11.40

14.40

14.40

14.40

15.00

15.00

15.00

43.20

43.20

43.20

EL

9.00

0.42

14.20

6.30

3.81

26.50

0.23

43.40

23.00

6.13

52.40

0.21

69.80

25.80

10.88

37.20

0.26

45.70

24.30

9.52

ES

21.90

0.29

32.10

9.40

6.24

22.10

0.21

29.20

13.00

4.61

53.20

0.13

67.50

44.40

6.73

42.30

0.21

58.50

22.60

9.04

FI

9.50

0.23

12.40

7.90

2.16

13.80

0.09

14.90

12.30

1.19

20.50

0.10

22.90

18.30

2.07

45.30

0.14

53.10

38.00

6.54

FR

9.00

0.26

15.00

4.20

2.32

15.20

0.53

41.40

9.80

8.06

24.30

0.43

56.30

19.70

10.54

43.70

0.17

60.50

31.20

7.29

HR

2.70

0.21

3.00

2.20

0.57

25.50

0.02

26.00

25.30

0.49

45.50

0.02

46.50

45.00

1.06

32.20

0.04

32.70

31.10

1.13

HU

11.40

0.35

18.40

7.20

3.96

17.40

0.25

24.10

13.60

4.31

20.40

0.30

28.90

12.80

6.05

34.10

0.25

48.70

25.40

8.48

IE

6.90

0.35

9.40

6.00

2.40

19.50

0.28

25.30

17.60

5.44

23.90

0.19

28.70

22.30

4.53

52.20

0.12

54.20

45.40

6.22

IT

15.00

0.30

24.00

8.40

4.44

29.00

0.28

42.10

11.70

7.99

42.70

0.29

59.70

12.40

12.53

23.90

0.17

31.60

17.40

4.15

LT

5.90

5.90

5.90

13.40

13.40

13.40

19.30

19.30

19.30

53.30

53.30

53.30

LU

6.10

6.10

6.10

8.30

8.30

8.30

22.60

22.60

22.60

52.70

52.70

52.70

LV

8.50

8.50

8.50

15.40

15.40

15.40

19.60

19.60

19.60

39.90

39.90

39.90

MT

20.30

20.30

20.30

10.40

10.40

10.40

11.70

11.70

11.70

26.50

26.50

26.50

NL

8.70

0.28

14.20

6.00

2.45

7.40

0.30

11.20

5.10

2.21

12.70

0.24

21.20

9.00

3.10

44.80

0.20

58.50

28.60

9.06

PL

5.40

0.46

10.70

3.30

2.46

16.00

0.22

24.20

11.80

3.48

23.90

0.24

41.10

17.70

5.68

42.10

0.14

56.60

28.70

5.89

PT

17.40

0.37

32.80

14.00

6.38

17.10

0.41

32.30

11.90

7.09

34.80

0.21

50.50

28.20

7.40

31.30

0.21

40.10

23.70

6.49

RO

18.10

0.31

25.00

9.00

5.55

21.40

0.33

31.80

14.00

7.12

24.00

0.34

34.00

12.40

8.22

25.00

0.39

47.60

17.60

9.77

SE

6.70

0.10

7.90

5.60

0.67

9.40

0.13

12.10

8.00

1.21

22.90

0.10

26.10

19.40

2.29

49.90

0.14

58.00

38.00

6.97

SI

4.40

12.00

0.34

14.90

9.20

4.03

20.20

0.28

24.00

15.90

5.73

41.00

SK

6.70

0.32

9.10

4.80

2.15

16.60

0.30

19.40

8.60

5.03

29.70

0.31

34.80

14.70

9.23

26.90

0.58

54.00

20.20

15.68

UK

11.80

0.27

19.50

7.20

3.16

15.60

0.22

23.10

8.90

3.42

16.90

0.24

26.80

8.60

4.08

47.70

0.15

60.70

31.60

7.28

Note: ESL: early school-leavers; NEET: not in employment, education and training; CV: coefficient of variation; MAX: maximum value; MIN: minimum value; SD: standard deviation; GDP: Gross Domestic Product; ISCED: International Standard Classification of Education.

(p.196)

(p.197)

Table 9.A2: List of variables

Variables

Category

Dimension

GDP at current market prices, euro per inhabitant

GDP and economic growth

ECONOMICS

Total intramural R&D expenditure in all sectors

Innovation

ECONOMICS

Researchers in all sectors as a % of total employment

Innovation

ECONOMICS

Motorways, total line, (1000/km2)

Infrastructure asset

ECONOMICS

Railways, total line, (1000/km2)

Infrastructure asset

ECONOMICS

Old dependency ratio, 2nd variant (65+ to population 15-64)

Pop. structure

DEMOGRAPHY

Median age of the population

Pop. structure

DEMOGRAPHY

Students at ISCED 5-6 as a percentage of population 20-24 years

Access

EDUCATION AND TRAINING

Students at ISCED 0-6 in all levels of education % of total population

Access

EDUCATION AND TRAINING

Students aged 17 (all ISCED levels) % of corresponding age population

Access

EDUCATION AND TRAINING

Early leavers from education and training (18-24 years), %

Transition from education to employment

EDUCATION AND TRAINING

NEET (people aged 15-24), %

Transition from education to employment

EDUCATION AND TRAINING

NEET (people aged 18-24), %

Transition from education to employment

EDUCATION AND TRAINING

Population (25-64) with ISCED 3-4, %, total

Upp. secondary attainment

EDUCATION AND TRAINING

Population (30-34) with ISCED 0-2, %, total

Primary and secondary attainment

EDUCATION AND TRAINING

Population (25-64) with ISCED 0-2, %, total

Primary and secondary attainment

EDUCATION AND TRAINING

Population (25-64) with ISCED 3-4, %, total

Upp. secondary attainment

EDUCATION AND TRAINING

Population with ISCED 3-4 (30-34 years), total %

Upp. secondary attainment

EDUCATION AND TRAINING

Population (25-64) with ISCED 5-8, %, total

Tertiary attainment

EDUCATION AND TRAINING

Population with ISCED 5-8 (30-34 years), total %

Tertiary attainment

EDUCATION AND TRAINING

Employment rate (25-54)

Employment

LABOUR MARKET

Employment rate (15-24)

Employment

LABOUR MARKET

Employment rate since education completion (five years), 20-34 years

Employment

LABOUR MARKET

Weekly hours of work in main job, 15-24 years

Typology of employment

LABOUR MARKET

Weekly hours of work in main job, 25-64 years

Typology of employment

LABOUR MARKET

Disposable income, net. PPS based on final consumption, per inhabitant

Disposable income

MATERIAL CONDITIONS

Population at risk of poverty or social exclusion, %

Poverty

MATERIAL CONDITIONS

At risk of poverty rate, % of population

Poverty

MATERIAL CONDITIONS

Severe material deprivation rate

Poverty

MATERIAL CONDITIONS

Life expectancy in age, more than a year

Life expectancy

HEALTH

Infant mortality rate

Infant mortality

HEALTH

(p.198)

Notes:

(1) We would like to thank Professor Yuri Kazepov for his collaborative work in the discussion of the theoretical framework for the quantitative analysis of Work Package 4 of the YOUNG_ADULLLT project.

(2) Detailed information about the territorial division of the European territory can be found at https://ec.europa.eu/eurostat/web/nuts/background. According to Eurostat, NUTS 1 corresponds to major socioeconomic regions; NUTS 2 are basic regions for the application of regional policies; and NUTS 3 are small regions for specific diagnoses, generally metropolitan areas.

(p.190) (3) As a matter of clarification, we are considering territorial unit as defined by the Eurostat classification system as explained in the previous section.

(4) The coefficient of variation is a measure of general entropy, which represents the variability in relation to the mean of the population. It is also known as the relative standard deviation.