To what extent are reforms made under external constraint maintained in the medium term? Under which conditions do decisions-makers reverse those, or alternatively keep them? This articles addresses these questions in Spain and Portugal. It shows that since 2014, 44% of the most important measures adopted under conditionality in both countries have been reverted. Nevertheless, the bulk (90%) of structural reforms enacted during the crisis remained. We also show that the left reverted more than the right; and that the preferences of veto players matter. Using crisp-set qualitative-comparative analysis (QCA), we find that within less than five years all temporary reforms were changed back, and that if businessor employers lobby for a reversal (and the costs to society are not salient), reversals will occur. Another finding is that most reversals are those with the largest electoral payoffs, as (cyclical) reversals always occur when the benefits are visible and concentrated rather than diffuse.