In the last decades, in most of the OECD countries employment relations have been changing and atypical forms of employment have been spreading rapidly. The “standard contract”, permanent and full-time, has lost importance, replaced by “flexible jobs”, such as fixed-term contracts, temporary work agency employment, variable working hours jobs, on call employment. A debate has been rising on whether atypical jobs, especially short-term contracts, might spur the development of a secondary labour market, in which the unemployed might get trapped in a cycle between dead-end jobs and unemployment. On the other hand, disadvantaged groups excluded from employment by too strict regulations might benefit most from the enhanced flexibility. Chapter 1 introduces this debate, Chapters 2 and 3 shed light on it. I analyse the labour market performance of workers who left unemployment through short-term jobs. I infer what counterfactual labour market performance would have been undertaken if the unemployed had rejected these jobs. In this way, it can be established whether short-term jobs may increase or decrease the chances of having a more stable career later in life, i.e. whether they are “stepping stones” or “dead ends”. I find evidence supporting the stepping stone hypothesis both in Italy and in Belgium. In terms of future job stability, even precarious and unsuccessful jobs are to be preferred to longer searches for directly finding better jobs. Chapter 4 is an identification analysis of the econometric models for duration data that encompass competing risks of exits, consecutive spells, and lagged duration dependence. Finally, Chapter 5 provides a new estimation strategy to look at the effect of past labour market experiences on two aspects of the subsequent job quality: wage and tenure on the job. The methodological novelty consists in jointly modelling labour market durations, transitions, and wages by way of a hazard-function based approach.