A vast literature has tried to assess the magnitude of the differences between wages in the industry and to explain why they exist. However, the measurement of inter-industry wage differentials and the consequent use of this information to assess the empirical relevance of different labor market theories have been hampered by the fact that total compensation measures - as opposed to wages And wages-, are not available in traditionally used data sets. In our view, we are the first to use compensation micro-data in a study of wage differentials between industries. Because non-wage compensation can easily exceed 40-50 percent of wages, its inclusion has the potential to alter the industry's measurable wage differentials by either decreasing or amplifying them. We found that the inclusion of benefits increases the dispersion of industry, measured by the standard deviation of the differentials between industries, by 16 percent when controls are not included and by an even greater 30 percent when controls are included.