Unemployment has been robustly shown to strongly decrease subjective wellbeing. Using panel quantile regression techniques, we analyse to what extent the negative impact of unemployment varies along the (conditional) subjective wellbeing distribution. In our analysis of British Household Panel Survey data (1996-2008), we find that individuals with high life satisfaction suffer less from becoming unemployed. A similar but stronger effect is found for a broad mental wellbeing variable (GHQ-12). Higher wellbeing seems to act like a safety net when becoming unemployed. We explore these findings by examining the heterogeneous unemployment effects over the conditional quantiles of various life domain satisfactions.