This paper studies individual log real wage change of east and west Germans over the period 1990/91 to 1995/96 using data from the German Socio-economic Panel. Parameters are estimated by median regression. Human capital theory and job mobility theories have implications for wage changes that differ between a transition and a non-transition economy. Wage changes decrease in experience according to human capital theory. But during a transition shock the introduction of new production techniques implies that experienced and inexperienced workers invest similarly in on-the-job training. Hence, annual wage changes do not vary much with experience. The estimation results reflect this difference. Matching and on-the-job search theory imply the following: A higher a worker's experience the higher his/her match quality and in turn the less (more) he/she can gain (lose) from an additional voluntary (involuntary) job change. But a transition shock destroys the high match quality of many experienced workers' jobs, so that wage changes due to voluntary and involuntary job change vary less with experience in a transition country than in a non-transition country. For involuntary job change, I find evidence for this hypothesis. Causal effects of job mobility on wages are estimated by comparing wage changes of this period movers with wage changes of next period movers. The results show that voluntary job changes raise real wages of both east and west Germans; involuntary changes tend to lower real hourly wages.