This article analyzes the microeconomic relation between innovation and employment, using company data from R&D Scoreboard for Europe covering 2000–2008. A reduced form equation in which R&D can account for both product and process innovation is estimated. The existence of non-constant elasticities is assessed, due to the combination of efficient scale and decreasing return to R&D: in our empirical estimates the scale effect tends to prevail for a given R&D intensiTY generating an increasing relation between total turnover and employment. These results have important implications for policymakers: innovation supporting policies should be correctly tailored and monitored since the results depend on the characteristics of the benefiting firms. Moreover, R&D intensiTY on GDP should be managed with care if taken as a policy target, given that the denominator is endogenous and non-linearly dependent on research expenditure.