We provide empirical support for the contention that within-job wage growth relates purely to job-specific performance and that returns to general experience are assessed at the point of job change. Using the British New Earnings Survey panel data we identify job changes that take place both within and between firms. We follow a cohort of 6778 male workers (born between 1958 and 1962) for the period 1975 to 1994 and estimate their within-job wage changes by allowing returns to vary over jobs. Between job changes are observed for individuals changing jobs in 1994. This work leads us to question the meaningfulness of attempts to separate returns to general and firm-specific human capital.