On the contribution of innovation to multi-factor productivity growth

We embed the innovation production function in a model that analyzes the impact of innovation output on manufacturing multi-factor productivity (MFP) growth. We combine a market share model with a gross output production function. This revenue approach enables a 'demand-shift' interpretation of the contribution of innovation to MFP growth. We apply different sets of instrumental variables and different estimation methods to estimate simultaneously the returns from innovation investment to innovation output, the contribution of innovation output to productivity growth and the feedback link running from a firm's overall sales performance to its innovation endeavor. We draw our empirical results from the second Community Innovation Survey (CIS-2) for the Netherlands. The estimation results from our model show that the impact of innovation differs between measures of firm performance, and that, in our data, the revenue function approach yields more sensible results for the contribution of innovation to MFP growth than the value-added production function framework. Furthermore, the results show that the estimation of return on innovation investment benefits from the inclusion of more information on the technological environment of the firm.