This study finds empirical support for the view that the Japanese bonu
s system represents, to a significant extent, returns to job-specific
tenure. The bonus also appears to play a significantly different role
from regular wage payments. An attempt is made to overcome some of the
difficulties in using the annual cross section data of the Basic Surv
ey of Wage Structure, an important data source of earlier studies. A p
seudo-panel is constructed which tracks given age cohorts through thei
r years of general experience and firm-specific tenure between the yea
rs 1971 and 1998. There is evidence of systematic changes in compensat
ion growth in relation to specific tenure among age cohorts. (C) 1995
Academic Press, Inc.