What drives youth unemployment?

What drives youth unemployment? 2017-08-29T16:10:36+00:00

The Great Recession of the late 2000s and ensuing economic recovery hit young people in the EU disproportionately hard, although with significant country variations. However, even before this major crisis, the performance of youth labour markets across the EU was highly varied and closely linked to the different institutional configurations at member state level.

Labour market performance: A single index measure

Here, we examine labour market performance affecting young people in the light of recent policies in Europe, drawing on an analysis of EU Labour Force Survey (EU-LFS) data in the period 2004-2012. Indeed, crucial to our analysis presented here has been a consistent availability of EU-LFS data, so that a time series of the EU before, during and after the recession can be constructed for the key age groups of young people: 15-19 and 20-24 year-olds.

Our aim was to develop a single index measure of labour market performance that would combine nine variables of labour market inclusion, human capital formation, labour market segmentation and transitions out of education.

The index provides a way of comparing the relative performance of youth labour markets across member states. It seeks to measure each country’s youth labour market performance (and achievement) on the basis of four key dimensions: (i) labour market inclusion; (ii) human capital formation; (iii) labour market segmentation; and (iv) school-to-work (STW) transitions, including transitions out of education and/or into NEET status.

Figure 1 below graphically presents the nine indicators used for this analysis that are, in turn, grouped in these key four dimensions of youth labour market performance. Crucially, these dimensions are closely related to strategic policy and programme objectives at both EU and member-state levels.

These aim to improve

  1. inclusion in the labour market and
  2. human capital accumulation while reducing
  3. segmentation and
  4. transitions from school to NEET status.
Figure 1: Framework of Performance Indicators

Source: Speckesser, S, (2015). D3.1 – Key Indicators and Drivers of Youth Unemployment, presentation at the STYLE Grenoble meeting, 5/3/2015

By adopting such an analytical framework and developing an index of multiple performance variables in order to assess country performance over time and across the EU, we sought to extend and enrich existing descriptive comparative studies of youth labour markets in two ways:

  1. Based on consistent data and formulas for 27 EU member states and a time series of nine years, our analysis is less affected by measurement issues, as cross-country differences in the multi-dimensional nature of policy performance over time are likely to be constant, similar to a fixed effect. In addition, to address measurement issues we used Principal Component Analysis to create a second index aimed at capturing a latent concept of labour market performance not affected by institutional differences in outcome variables. For example, by doing so we sought to address the issue of possibly underestimating youth unemployment in countries with large-scale apprenticeship systems.
  2. For individual countries, we describe how performance improved over time by indexing the starting value in 2004 to 1. This allows us to show whether a country’s combined performance along the four key dimensions mentioned above improved over time.

This study does not follow the conventional conceptual framework of econometrics, which aims to obtain estimates of the quantitative magnitude of effects of institutional reform. In this case, one is, for example, interested in estimating how much macroeconomic outcomes − such as high employment levels or economic growth − would change if particular institutional arrangements were altered (Hadjivassiliou et al. 2015; Gonzalez Carreras et al. 2015; Hadjivassiliou et al. forthcoming). In contrast, here our focus is on the description of performance and measures associated with youth-related policy reform in particular countries. In doing so, we adopt an inductive rather than a theory-led approach to obtain high-level inference on how institutional change can affect complex outcomes, such as young people’s transitions to the labour market that are, in turn, affected by a multiplicity of factors such as those examined here.

Nine indicators

To this end, as mentioned earlier, we first used the nine indicators (shown in Figure 1 above) to describe each country’s performance across four dimensions: (i) labour market inclusion; (ii) human capital formation; (iii) labour market segmentation; and (iv) STW transitions.

For example, in terms of employment rates – a key indicator of labour market inclusion – Denmark, Norway, the Netherlands and the German-speaking countries are the best performers with their high and stable youth employment rates over time.

Likewise, in terms of STW transitions as measured by a young person’s transition into NEET and unemployed status, the Netherlands and the German-speaking countries outperformed all the other member states over the entire period under study (2004-2012). These were closely followed by the Nordic countries, which also managed to reduce the proportion of young people becoming NEET/unemployed after leaving education.

Policy effectiveness

We have additionally created index measures for the analysis of policy effectiveness in the light of the strategic objectives set at EU and national levels. In doing so, we sought to address the multi-dimensional nature of policy performance since index measures are able to capture improvements in some dimensions, such as a reduction of long-term unemployment, while other dimensions of youth labour market performance remain unchanged.

In cross-country analyses such as the one presented here, using index measures to compare country performance can both reveal how much countries have improved in relation to the best performer or an absolute benchmark (‘aggregate loss function’) and also show whether differences in youth labour market performance narrowed or widened over time. On the basis of such index-based analysis we are, in turn, able to both examine a country’s youth labour market performance over time and show whether policy performance has improved.

Composite indicators

To this end, two composite indicators were constructed summarising each country’s performance over time and in relation to the two groups of young people under study (i.e., those aged 15-19 and those aged 20-24). Overall, on the basis of the composite indicators, Denmark, Germany and the Netherlands are the best performers in terms of youth labour markets. This is, broadly, in line with the strong body of evidence that consistently shows that countries with effective dual (work-based/apprenticeships) or school-based VET systems are characterised by successful STW transitions.

A third part of our analysis involved using the composite indices to compare change in a country’s youth labour market performance over time. This showed that Austria, Belgium, Poland and the Nordic countries improved in relation to countries with the best youth labour market performance between 2004 and 2012. However, the gap between Southern European countries and the top performers has widened, reflecting both structural factors such as the weak links between education and the labour market, in the former, as well as the fact that they were hit disproportionately hard by the Great Recession and associated Eurozone crisis.

A fourth part of our analysis comprised the estimation of probit models using EU-LFS micro-data to identify the potential drivers of and barriers to STW transitions and how these have changed over the last ten years. The probit models show that, in terms of individual characteristics, the main barriers to effective STW transition include the following: (i) being young; (ii) being female; (iii) having low levels of educational attainment (and no or low qualifications); and (iv) having an immigration background/foreign nationality. On the other hand, higher levels of qualifications and parental qualifications definitely improve transitions from education to employment.

NEETs

Again, our findings are in line with the existing evidence base according to which the level of educational attainment is a key determinant of young people’s likelihood of being unemployed and/or NEET. For example, as the seminal work of Eurofound (2012) has shown, young people with low educational attainment (ISCED 0-2) are 300% more likely to be NEET than better qualified youth, while those with an immigration background are 70% more likely to become NEET.

STW transitions are becoming relatively more segmented

In general, we assumed that where the barriers to youth labour market integration have increased over time (i.e., between 2004 and 2012), STW transitions are becoming relatively more segmented in relation to people’s existing qualifications, gender, nationality or the effects of parental education on youth transitions.

In contrast, where barriers have decreased, then STW transitions have been relatively less segmented. Indeed, our analysis also showed that, over time, the significant barriers of gender, age, foreign nationality and education levels to successful STW transitions decreased in most EU countries. Not surprisingly, consistent with our earlier labour market performance assessment, the individual barriers to STW transitions decreased most in countries with improving youth labour market conditions. That said, as is always the case with EU-wide comparative analysis, the emerging picture is more nuanced. For example, in some countries, such as Austria, Belgium and France, young people with low educational attainment levels now have more difficulties than before the Great Recession.

Moreover, we found no clear picture on how an overall deteriorating economic situation affected the individual transitions and labour market performance of a particular country. For example, in Portugal and Spain, where the youth labour market deteriorated dramatically as a result of the Great Recession and its aftermath, the impact of individual characteristics such as gender became less important altogether.

Gender differences

However, such ‘fading gender differences’ cannot be interpreted as progress for young women in terms of achieving better labour market outcomes; rather, the recession had a stronger negative effect on young men’s STW transitions. For example, the collapse of the construction sector in Spain, which historically employed a large share of young men, resulted in a 72% fall in youth employment between 2008 and 2011. In other countries, drivers and barriers to STW transitions have remained broadly unchanged.

Parental education

Finally, parental education was also found to significantly affect young people’s STW transitions. Young people whose parents have low educational attainment (ISCED 0-2) are up to 150% more likely to be NEET compare to those whose parents have a secondary level of education, and are up to 200% more likely to be NEET than those whose parents are tertiary education graduates (Eurofound 2012).

Growing inequality in young people’s transitions

This, in turn, suggests growing inequality in young people’s transitions originating from family circumstances in the form of parental level of educational attainment. Indeed, this could be a further issue for policy action in addition to improving the labour market situation for young people more generally.

Overall, the use of our composite indicators has exposed a latent concept of youth labour market outcomes. This, we argue, provides a more useful tool for monitoring policy progress and for evaluating and interpreting the outcomes for high-level policy decision-making in relation to STW transitions.

Moreover, the analysis of aggregate indicators needs to be complemented by an analysis of micro-data on individual drivers of and barriers to successful transitions so as to improve the targeting of policy to specific groups and decrease inequality.

References

Eurofound. 2012. NEETs − Young People not in Employment, Education or Training: Characteristics, Costs and Policy Responses in Europe, 21.10.2012, http://www.eurofound.europa.eu/sites/default/files/ef_publication/field_ef_document/ef1254en.pdf

Gonzalez Carreras, Francisco, Laura Kirchner Sala and Stefan Speckesser. 2015. The Effectiveness of Policies to Combat Youth Unemployment. STYLE Working Paper WP3.2 Policies to combat Youth Unemployment

Hadjivassiliou, Kari, Laura Kirchner Sala and Stefan Speckesser. 2015. Key Indicators and Drivers of Youth Unemployment. STYLE Working Paper WP3.1 Indicators and Drivers of Youth Unemployment

Hadjivassiliou, Kari P., Arianna Tassinari, Werner Eichhorst and Florian Wozny. Forthcoming. ‘How does the performance of school-to-work transition regimes in the European Union vary?’ In Youth Labor in Transition, edited by Jacqueline O’Reilly, Janine Leschke, Renate Ortlieb, Martin Seeleib-Kaiser and Paola Villa. New York: Oxford University Press.