In 2015, no less than 15 per cent of young people aged 15-29 in France were Not in Employment, Education or Training (so-called "NEETs"). This young population is believed to be a homogeneously vulnerable group, risking detrimental long-term socioeconomic marginalization. NEETs represent one of challenging age groups with regard to the level of labour market participation. NEETs are an essential problem with huge economic and social costs. Despite these immediate concerns, the theoretical understanding of NEETs is still limited. Little is known about whether NEET is a transitory or long-lasting period, about differences in the timing (age) of becoming NEET (15-19, 20-24, and 25-29), and about the heterogeneity of the NEET population. Moreover, the role of institutional contexts is not well documented. The aim of this paper is to provide theoretical and empirical insights on the temporal patterns of NEET. First, we focus on the impact of individual and institutional characteristics in order to understand the risks of becoming NEET. Secondly, we address the effects of the interactions between institutions and individual characteristics on young people's transitions into and out of NEET. From a theoretical perspective, our paper will focus on labour segmentation theories and their application to the French labour market. Researches using this theoretical framework have frequently linked education level of youth entrants and barriers to entry in the stable segments. We complete this theoretical approach by researches on social stratification of the French educational system: participation in upper secondary education is strongly related to the socioeconomic background, immigration history etc. A social and educational exclusion mechanism leads to a marginalization of high school dropouts, specifically from students who had been enrolled in a vocational track. Beyond these theoretical explanations, the economic context might influence NEET risks and consequences. The economic recession characterised by the lacks of jobs for young entrants may reinforce the effect of level of education in the job queue. We use French Continuous Labor Force Survey carried out by INSEE that provides information of labour situation and transition. First, cross-sectional data are used to understand NEET patterns, to assess effect heterogeneity between individuals and reveal underlying mechanisms of NEET risks. Second, longitudinal data are used to analyse individual and institutional effects on trajectories in and out of NEET. More precisely, we will focus on labour market transition patterns just before and after the crisis. Using Kelly et McGuiness (2015) methodology, we will test if the factors associated with being NEET have change since the economic crisis. Our preliminary econometric results confirm the heterogeneity of the NEET population and show how the NEET risks vary according to age, sex, educational level and social origins. In addition, focusing on the young people who responded to the first and the fifth questionnaires, our results provide also some evidence that a NEET transition towards an employment is more related to the educational attainment, after 2008.