This paper examines the impact of innovations and wages on the demand for heterogeneous labour. Based on matched data from the IAB-establishment panel survey and the files of the employment statistics register for the year 1995, input shares derived from a generalised Leontief cost function are estimated for six qualification groups (blue and white collar workers stratified into unskilled, skilled and highly-skilled employees) in the West German production industries. With the exception of highly skilled blue collar workers, innovations have a positive and significant effect on labour demand for all groups, with an estimated relative change of the conditional labour demand ranging between 3.2% and 6.3%. Between white collar workers as well as between unskilled and skilled blue collar workers, we find an increasing positive impact of innovations on labour demand with qualification level. Skilled and highly skilled employees are found to be substitutes for the unskilled. This result implies that more flexible wages of the unskilled would reduce the unemployment of this group. Finally, our data is not consistent with separability of blue-collar employees from white collar employees, there is weak evidence that skilled and highly skilled blue collar employees can be aggregated.