The effectiveness of conservation interventions such as Payments for Environmental Services (PES) is often evaluated--if it is evaluated at all--only at the completion of the intervention. Since gains achieved by the intervention may be lost after it ends, even apparently successful interventions may not result in long-term conservation benefits, a problem known as that of permanence. This paper uses a unique dataset to examine the permanence of land use change induced by a short-term, asset-building PES program implemented in Quindío, Colombia, between 2003 and 2008. This the first PES program to have a control group for comparison. Under this program, PES had been found to have a positive and highly significant impact on land use. To assess the long-term permanence of these changes, both PES recipients and control households were re-surveyed in 2011, four years after the last payment was made. We find that the land use changes that had been induced by PES were broadly sustained in intervening years, with minor differences across specific practices and sub-groups of participants, indicating that these changes were in fact permanent. The patterns of change in the period after the PES program was completed also help better understand the reasons for the program's success. These results suggest that, at least in the case of productive land uses such as silvopastoral practices under conditions such as those at the study site, asset-building PES programs can be effective at encouraging land owners to adopt environmentally-beneficial land management practices and that the benefits will persist after payments cease.