Final Report Summary - MADEM (Market Design and the Evolution of Markets)
MaDEM builds on the premise that markets “do not fall from sky” (like the Walrasian auctioneer in economic textbooks) but are created, their rules evolve in response to their environments and participants learn to use them. It aims to further our understanding of how markets get created, how they evolve and how their design impacts performance. This was done through the analysis of three distinct sets of markets: financial markets for derivatives, carbon markets, and allocation markets without financial transfers. In each case, new data were collected and new insights have been generated from the combination of theory and data analysis.
Understanding competition between financial markets
A long-standing puzzle in finance concerns the reasons why the same assets are sometimes traded in different trading venues even though traders value liquidity. Existing explanations have largely focused on trading decisions. Our research has documented that exchange membership decisions, which are decisions that take place before trading decisions, are a large part of the explanation and we have quantified for the first time the relative importance of factors driving membership. In ongoing work, we are studying switching costs due to clearing and how these may vary across assets. This line of research contributes to a better understanding of the impact and limits of competition between financial exchanges at a time when this industry is undergoing a wave of entry and consolidation.
Market creation and market development
We are exploiting recently released data from the first phase of the EU emissions trading scheme to understand how market design impacts its development and, eventually, its performance. We know who traded, when, with whom, on which platform if any, and at what price. These unusual data allow us to map the actual market structure that emerged. We then build a model that integrates the salient features of the data. The model serves to explain why allowance prices in the EU ETS may have failed to properly aggregate information and how market design, including rules on auctioning and banking, can help improve information aggregation.
Markets without prices
When there is no price to balance supply and demand, other rules must be used to allocate goods that are in scare supply. Traditional criteria that guide this choice include efficiency of the resulting allocation and strategyproofness. In this part of MaDEM, we have studied two allocation problems: how to allocate seats in courses to students when there is a cap on the number of students in a course, and how to assign students to schools. The existing literature had so far advocated dictatorship as the best solution to the course allocation problem on the basis of its strategyproofness. Using data from Harvard Business School, we have quantified the large negative welfare impact of such rule, uncovered a new link between fairness and efficiency and proposed a new rule that largely outperforms dictatorship. In ongoing work, we are analyzing how allocation rules interact with preferences in exacerbating or reducing preference polarization when students take their likely peers into account when forming preferences over schools. A policy change in the city of Ghent (Belgium) provides a test for the predictions of the theory.
The grant has produced 9 published (or in R&R stage) articles, 2 articles in conference proceedings, 4 working papers and 4 other papers in preparation. It has funded fully or partially 4 post-docs and provided seed-financing for 6 doctoral students. It has sponsored the creation of the European-based research network “Matching in Practice” that was set to foster research on matching for education markets and its dissemination. Research on school choice sponsored by MaDEM was the basis of numerous dissemination activities in the PI’s home country.
Understanding competition between financial markets
A long-standing puzzle in finance concerns the reasons why the same assets are sometimes traded in different trading venues even though traders value liquidity. Existing explanations have largely focused on trading decisions. Our research has documented that exchange membership decisions, which are decisions that take place before trading decisions, are a large part of the explanation and we have quantified for the first time the relative importance of factors driving membership. In ongoing work, we are studying switching costs due to clearing and how these may vary across assets. This line of research contributes to a better understanding of the impact and limits of competition between financial exchanges at a time when this industry is undergoing a wave of entry and consolidation.
Market creation and market development
We are exploiting recently released data from the first phase of the EU emissions trading scheme to understand how market design impacts its development and, eventually, its performance. We know who traded, when, with whom, on which platform if any, and at what price. These unusual data allow us to map the actual market structure that emerged. We then build a model that integrates the salient features of the data. The model serves to explain why allowance prices in the EU ETS may have failed to properly aggregate information and how market design, including rules on auctioning and banking, can help improve information aggregation.
Markets without prices
When there is no price to balance supply and demand, other rules must be used to allocate goods that are in scare supply. Traditional criteria that guide this choice include efficiency of the resulting allocation and strategyproofness. In this part of MaDEM, we have studied two allocation problems: how to allocate seats in courses to students when there is a cap on the number of students in a course, and how to assign students to schools. The existing literature had so far advocated dictatorship as the best solution to the course allocation problem on the basis of its strategyproofness. Using data from Harvard Business School, we have quantified the large negative welfare impact of such rule, uncovered a new link between fairness and efficiency and proposed a new rule that largely outperforms dictatorship. In ongoing work, we are analyzing how allocation rules interact with preferences in exacerbating or reducing preference polarization when students take their likely peers into account when forming preferences over schools. A policy change in the city of Ghent (Belgium) provides a test for the predictions of the theory.
The grant has produced 9 published (or in R&R stage) articles, 2 articles in conference proceedings, 4 working papers and 4 other papers in preparation. It has funded fully or partially 4 post-docs and provided seed-financing for 6 doctoral students. It has sponsored the creation of the European-based research network “Matching in Practice” that was set to foster research on matching for education markets and its dissemination. Research on school choice sponsored by MaDEM was the basis of numerous dissemination activities in the PI’s home country.