Restorative justice goals are frequently articulated on micro, meso and macro levels. One macro-level goal frequently made by advocates is that restorative justice may serve as a viable means of reducing incarceration. Focusing on Australia, New Zealand, the United Kingdom and the United States, this article argues that while these countries have seen some of the largest increases in incarceration within western industrialized countries, as well as the most widespread use of restorative justice, there is little evidence that restorative justice has reduced prison populations. It also argues that as currently practiced there is little reason to assume that restorative justice will have a significant impact on incarceration in the near future. Attention is given to the problem of the ‘transformation assumption’ inherent in restorative justice that micro-level changes in offender behaviours or restorative outcomes can significantly affect the larger social structures of punishment and incarceration.