Background: The literature has shown that the way active labour market policies (ALMP) aimed at youth are designed and implemented can influence the labour and educational prospects of youngsters. The evaluation of the Catalan PQPIs (initial vocational qualification programmes) presented here seeks to provide new evidence on the effectiveness of ALMPs for a social, economic and geographical context--the south of Europe--in which such evaluations are scarce but particularly necessary, given current unemployment conditions. Purpose: This evaluation focuses on the impact of the PQPIs on participants' labour and education outcomes during the period 2009-2013. Research questions are: Does participation in the PQPIs increase the probability of finding a job? What is the impact of the PQPIs on the probability of returning to formal education? Does the impact of the PQPIs vary according to characteristics of the participants, such as age and gender? Programme description: First launched in 2006, PQPIs target 16-25-year-olds who are unemployed and who left education without earning the certificate of compulsory secondary education (GESO). PQPIs have two objectives: (a) to increase the labour prospects of participants, and (b) to facilitate their re-engagement into formal education, either by completing the GESO or by enrolling in middle vocational courses (CFGM). PQPIs take one academic year to complete and they include vocational training, work internships and preparation for returning to formal education. Sample: Our evaluation focuses on the two main types of PQPI, the PQPI-FIAP and the PQPI-SBS. Our sample consists of all students who started any of these courses during the 2008-2009 academic year--2401 participants in the PQPI-SBS and 1220 in the PQPI-FIAP. Both sets of students were observed from the beginning of their courses until April 2013. Design and methods: Using propensity score matching, we estimate the intention-to-treat and treatment-on-the-treated impact of PQPIs on participants' labour and educational outcomes. For labour outcomes, we focus on the monthly labour force participation from the moment that the programmes started (October 2008) until April 2013. For educational outcomes, we analyse the accumulated percentages of GESO and CFGM enrolments and graduations from the time the programmes finished (June 2009) until September 2012. Results: Results show that both the PQPI-FIAP and the PQPI-SBS are ineffective in increasing the labour prospects of participants. In contrast, both the PQPI-FIAP and the PQPI-SBS are significantly effective in re-engaging youths in formal education, although such gains are not translated into a higher level of attainment in post-compulsory education. The positive educational impact identified is concentrated among 16-18-year-olds, while the effects are null or even negative for all other age ranges. Conclusions: First, the study demonstrates the shortcomings of ALMPs and school-to-work transition programmes in increasing the labour participation of their participants. Second, despite these shortcomings, PQPIs do function as a mechanism for re-engaging early school leavers back into formal education. Third, the educational effectiveness of programmes such as the PQPIs question the capacity of comprehensive schooling to prevent early school leaving, especially when comprehensiveness is implemented without any margin of flexibility.