Heterogeneous Technological Changes and Their Impacts on Income Inequality and Productivity Growth
Technological Change , Income Inequality , Productivity Growth
I investigate different types of technological change and their impacts on wage inequality and economic growth. In particular, I distinguish between skill-complementing and skill-replacing technological changes, with an emphasis on the latter. Skill-replacing technological change can take the form of an organizational change, or the development of a certain type of machinery that works with lower skilled workers. Essentially, skill-replacing technological change increases specialization in tasks, so that less labour skill is required. In contrast, skill-complementing technological change tends to increase the productivity of and demand for high-skilled workers. Combining two novel European data sets, I first document a previously unknown fact: industries with proportionally more product innovation relative to process innovation tend to pay higher skill premia, as measured by the wages of college educated workers relative to those of high school graduates. To further understand this observation, I develop a dynamic general equilibrium model in which firms conduct both product innovation and process innovation endogenously. This model provides a stylized interpretation in which product innovation is skill-complementing, and process innovation is skill-replacing. I show how each type of innovation affects the skill premium through the labour demand channel, and also how the skill premium affects the incentives to develop each type of innovation. Using the European data, I estimate this two-way relationship derived from the model at the industry level. I address the issue of simultaneity using an instrumental variables approach suggested by the model. I also calibrate the model and conduct some quantitative exercises. Finally, I develop an endogenous growth framework featuring minimum skill requirements, two independent process innovations, and a heterogeneous labour force. This model can simultaneously account for the hollowing out of the middle class and the slow down in labour productivity growth, both of which were observed in the United States and most European countries in the recent past. I fully characterize the equilibrium of the model and derive conditions for both phenomena to emerge. The relative intensity of different types of technological change plays a central role in the framework.