This paper explores the effect of time varying velocity in a transition to price stability. Nonstationary velocity, expressed as function of consumption, is made endogenous in Ireland’s (1997) model. We find that the ‘disinflationary booms’ found by Ball (1994) may or may not disappear; and also that temporary output losses may be much larger than previously thought, depending on velocity. A gradual disinflation of low inflation may even be undesirable given its overall negative impact on the economy. Finally, we explore the optimal speed of disinflation.