The reservation wage is an integral part of most theories of involuntary unemployment. We use panel data to examine the empirical determinants of the reservation wage - in particular the inßuence of previous wages - and consider what this implies for the evolution of the natural rate of unemployment. We Þnd that previous wages have a signiÞcant but relatively small effect on reservation wages (an elasticity between 0.15 and 0.47). We also Þnd considerable differences across genders with previous wages being more important for men and market wages being more important for women. Overall our results suggest that unemployment will adjust relatively quickly to shocks.