Policies addressing economic effects of new automation technologies

The last decades have witnessed a rapid development and adoption of new automation technologies. These technologies transform the economy and, in particular, the labor market. Recent evidence suggests that automation is partly responsible for the decline in the labor share and the polarization of real wages. This cumulative dissertation is dedicated to the issue of this emerging income inequality, exploring the following policy instruments: a binding minimum wage, a task-specific tax, and policy complementarities to curb market power. This is done in three stand-alone papers, of which two papers have already been published. The first paper examines whether a minimum wage is a suitable policy instrument for reducing the emerging income inequality resulting from automation. For this aim, we introduce a binding minimum wage in a task-based framework, in which tasks are performed by machines, low-skill, and high-skill workers. We show that the inequality between the low-skill wage and the other factor prices decreases when the minimum wage increases. However, an increase in the minimum wage may create ripple effects where low-skill workers are displaced by the other production factors and, consequently, the income share of low-skill workers is nonincreasing. Further, we explore the effects of automation on labor market outcomes along five different margins. In the presence of a minimum wage, we show that the output effects of automation at the extensive margin and the creation of new tasks are ambiguous. The reason for this is twofold: on the one hand, a positive impact of the cost-saving effect and, on the other hand, a negative impact of the displacement of low-skill workers. In particular, we emphasize a potential trade-off between less inequality of the factor prices and greater inequality of the income shares. The second paper analyses the selection of the preferred level of automation in a political economy consisting of low-skill and high-skill workers. To control the level of automation, we implement a task-specific tax in a task-based framework. First, we assume that low-skill workers have the full power to select their favorite level of automation. In this setting, we show that they prefer to tax machines such that the level of automation is lower compared to the case without a tax. Then, we examine the impact of different types of technical changes and show that they influence the low-skill workers' selection of their favorite level of automation. Hence, these changes affect the inequality between low-skill and high-skill workers. Finally, we consider the high-skill workers' favorite level of automation and the aggregated outcome of a political process between both groups of workers. We observe that high-skill workers prefer a taxation of low-skill workers to increase the level of automation beyond the case without a tax. Therefore, the greater the power of high-skill workers in the political process is, the greater the selected level of automation is. The third paper explores whether broad reforms are more advantageous than singular policies for raising workers' real wages. Recent evidence suggests that - most likely due to economies of scale - automation has been accompanied by an increase in firms' market power, not only in the product market but also in the labor market. As a result, mark-ups have increased, and nominal wages have come under pressure. As anti-competitive behavior of firms reduces workers' real wages, policymakers usually focus on antitrust regulation but neglect to curb employers' monopsony power. We show that there is a policy complementary by addressing both types of market power. Hence, broad reforms should be preferred.