This article analyses change of time use and time allocation in three schools participating in a Swedish five-year national experiment in which State regulation of teaching time was weakened. Participating schools could freely decide how to use and distribute teaching time. The experiment was launched at a late stage in a 25-year decentralisation process. During this period, the Swedish education system has become one of the most decentralised ones among the OECD countries. Based on a four-year longitudinal study, the initiation and implementation of more goal-oriented and flexible time allocations in the three schools were analysed. When removing the time schedule one would expect schools to change both time allocation and pedagogy. However, in all three schools, change concerned the latter rather than distribution of teaching hours across subjects, pupils and so on. In particular, change was about replacing traditional subject-based teaching by thematic, cross-disciplinary studies and introducing working forms resulting in increased autonomy, but more responsibility on the part of the pupils. It is concluded that the additional weakening of time governance did not have any dramatic effects on initiation and implementation of school development. It primarily resulted in a confirmation, legitimisation and to some extent speeding up of existing change efforts. More generally it led to increased possibilities of information, networking and attention. (Contains 3 tables.)