We construct firm-level data set with matched productivity and qualification data by linking the Annual Business Inquiry and Employer Skills Survey for England. We first examine the effect of workplace skills and other characteristics such as part-time status and gender on both productivity and wages in English firms. We also investigate how productivity-implied returns to worker characteristics compare with wage-implied returns, therefore providing information on how rents are distributed between employers and employees. We find that firms with a higher share of college-educated, full-time and male workers also tend to be more productive, with considerable variations across sectors. The only robust difference in implied returns follows from part-timers, who tend to work for firms that pay too low wages for the observed productivity differences. Second, we study the effect of local skills on productivity controlling for skills at the firm. We find a positive and robust association, which is consistent with positive human capital externalities.