Authors: Herbert Dawid, Philipp Harting, and Michael Neugart
Abstract: Artificial intelligence algorithms are increasingly used for online pricing and are seen as a major threat to competitive markets. We show that if firms use a deep Q-network (DQN) as an example of a state-of-the-art machine learning algorithm, prices are supra-competitive in duopoly but quickly move to competitive prices as the number of competitors in an oligopoly increases. This finding is very robust concerning variations of the exploration and learning rate used in the DQN algorithm.
Author: Darius Griebenow
Abstract: The EU Emissions Trading System is the cornerstone of Europe's strategy to comply with the Kyoto Protocol and lower its greenhouse gas emissions sufficiently to meet climate goals. Since its implementation in 2005, researchers have sought to not only determine the program's effectiveness in reducing carbon emissions, but also the potentially adverse economic consequences for regulated entities. I study the economic performance of regulated and unregulated firms in Spain, a heavily credit-constrained country whose firms have operated in a deteriorated economic enviroment for a prolonged period of time. Using a difference-in-difference methodological approach, I find regulated Spanish firms benefitted from higher revenues, employment levels, and fixed asset stocks compared to their unregulated counterparts. I also investigate potential channels driving these effects, and find that access to debt is a particularly important factor to consider in the context of Spain. A closer look at the dominant ETS sector in Spain, ceramics and clay production, shows that despite fierce external and internal competition, the sector's ETS firms thrived in particular.
Authors: Darius Griebenow, Michael Neugart, and Stefan Pichler
Abstract: Policies aimed at reducing carbon dioxide emissions are crucial in the fight against climate change. We explore whether political parties that implement environmental regulations face electoral backlash, which could hinder or even obstruct the adoption of climate policies. In particular, using detailed data from the European Union Emissions Trading System, we construct a new index for the stringency of environmental regulation at the regional level (NUTS2) for the European Union. This index is used to assess the economic impact of heightened environmental regulation on regions and to analyze voter reactions to these policies at the polls. Our findings reveal that green parties benefit from stricter environmental regulations in national elections, while this effect is absent in European Parliament elections. The increased vote share for green parties seems to result directly from the regulation rather than being mediated by improvements in regional economic performance.
Authors: Darius Griebenow, Jens J. Krüger, and Michael Neugart
Abstract: When setting up the world's largest emission trading system to combat carbon dioxide pollution, European policymakers were facing strong pressure to supply industry with free allocations of emission allowances. This has led to the issue of over-allocation of allowances with, however, limited evidence to which extent this over-allocation occurred. After all, measuring the extent of over-allocation properly requires the construction of a meaningful counter-factual. We propose an approach which exploits a policy change in the allocation of allowances in 2013, when the third phase of the European Emissions Trading System (ETS) started. From then onward, electricity producers were not granted free allowances anymore, in contrast to all other installations which were endowed with allowances. By comparing these groups, we attempt to assess how installations would have behaved in Phase 3, had they still received free allowances as the comparison group did. Compared to Phase 2, electricity producers in Phase 3 were, on average, holding 505804 allowances less. This change in stocks is driven by the policy-induced reduction in the free allowances, while no change can be observed in purchases, sales, and carbon dioxide emissions.
Sponsored by the Federal Ministry of Education and Research
Digital Green Tech funding measure – environmental technology meets digitalization. Duration: 9/2023 to 9/2025.
Short summary: Water management in Germany is neither resource-efficient nor does it meet needs as needed: drought leads to supply gaps, operation is not energy-optimal and there are pipe losses. New, flexible, resource-saving (resources: water, energy, money), but also comprehensible operating strategies are necessary. Although data-based methods of artificial intelligence and heuristic approaches hold considerable potential, questions arise immediately regarding the availability of the necessary data as well as the traceability and transparency of the approaches. These are the reason for the approval trap of autonomous systems. However, if market economy mechanisms are used, with adequate market design, the market's inherent resource and allocation efficiency can be used to achieve comprehensible, resource-efficient and needs-based operating strategies.
Project partners: Michael Neugart, Peter Pelz, Rolf Findeisen, Wilo SE, RheinEnergie
Authors: Herbert Dawid, Philipp Harting, and Michael Neugart
Abstract: We study how the use of machine-learning based algorithms for the determination of wage offers affects workers’ wages on online labor platforms. Firms use reinforcement-learning to update posted wages on the platform, and heterogeneous workers send applications based on the posted information. We show that if firms use a deep Q-network (DQN), as an example of a state-of-the-art machine learning algorithm, the emerging wages closely resemble the equilibrium outcome. However, slightly changing the setup of the algorithms can lead to substantial collusion and wages well below the equilibrium level. In particular, we identify a specific property of the algorithms, namely whether experience replay is used, which determines whether collusion occurs or not. Our findings are robust with respect to many features of the model, including the design of the online labor platform.
