The low unemployment rates traditionally enjoyed by Sweden have often been attributed to the country’s extensive system of active labour market programmes, which have thus often been regarded as a model for other countries to emulate. The paper investigates the presence of short- and long-term treatment effects on a number of outcomes, including employment and unemployment benefit collection. Special attention is devoted to subsequent outcomes experienced by former participants, in particular to their job attachment and their performance should they fall back into unemployment. Finally, the distinctive feature of the Swedish labour market policy, whereby participation in programmes renews eligibility to generous unemployment compensation, is investigated in relation to the incentives it is likely to create to keep cycling between compensated unemployment spells and programme participation. The approach used is propensity score matching, with some additional analyses trying to account for a partially unobserved outcome variable due to misclassification problems in the data. Joining a programme as opposed to waiting longer in open unemployment appears on average to have a positive dynamic effect on participants’ employment rates. The overall findings indicate however that the human capital-enhancing component of the programmes may not always be strong enough to outcompete the work disincentives provided by the system. Furthermore, even when cycling has been ruled out by focusing on individuals observed to exit their unemployment spell, programmes are found to have no effect on any of the outcomes considered.