Calendrier du 16 novembre 2017
Macroeconomics Seminar
Du 16/11/2017 de 15:45 à 17:00
PSE - 48 boulevard Jourdan, 75014 Paris - salle R2-21
OIKONOMOU Rigas (UC Louvain)
*
TOM (Théorie, Organisation et Marchés) Lunch Seminar
Du 16/11/2017 de 12:30 à 13:30
salle R2-20, campus Jourdan, 48 bd Jourdan - 75014 Paris
MOULIN Hervé (University of Glasgow)
Fair Mixture of Public Outcomes
écrit avec joint with H. Aziz and A. Bogomolnaia
Abstract : We revisit probabilistic voting under dichotomous preferences where, as on Facebook, agents report that they like/dislike every outcome. We focus on two often conflicting normative concerns. Minorities should not be crushed by the opposing majority; and the size of the support for a given outcome should increase its weight: numbers matter. The Unanimous Fair Shares property adresses both concerns by giving to any group of like-minded agents an influence proportional to its size. Individual Fair Shares weakens UFS by only guaranteeing to each agent a 1/n-th influence on the outcome (where n is the total number of agents).
Given that we cannot combine Efficiency, Incentive Compatibility (Strate- gyproofness) and Unanimous, or even Individual, Fair Shares, we propose second best mechanisms achieving two of these three design goals.
The Conditional Utilitarian rule, a simple variant of the classic random dictator, is Strategyproof and guarantees Unanimous Fair Shares. It is much easier to compute and more efficient than the familiar Random Priority rule. In numerical simulations its inefficiency is consistently low.
The efficient Egalitarian rule guarantees Individual Fair Shares and is Excludable Strategyproof: this weakens Strategyproofness by ensuring that an agent is excluded from consuming those public outcomes she reportedly dislikes. But numbers do not matter: the rule treats a unanimous group of agents exactly as if it contains a single agent.
The efficient Nash rule (maximizing the product of utilities) offers much stronger welfare guarantees than both above, in particular it achieves core stability if any group of agents can enforce any outcome with a probability proportional to its size. But the rule fails even the excludable form of Strategyproofness.
We also uncover several challenging open questions
Texte intégral
Travail et économie publique externe
Du 16/11/2017 de 12:30 à 13:45
MAGNAC Thierry (Toulouse School of Economics)
A Pigouvian Approach to Congestion in Matching Markets
Matching markets often require recruiting agents, "programs," to costly screen "applicants" who are agents on the other side. A market is congested if programs have to screen too many applicants. A cost associated with application submission is a Pigouvian tax to mitigate the negative externality imposed on programs by applicants. A higher cost reduces congestion by discouraging applicants from applying to some programs, which may, however, put match quality in jeopardy. We measure the effects of such Pigouvian taxes by studying variants of the Gale-Shapley Deferred-Acceptance mechanism with differential application costs. Using data collected in a multiple-elicitation experiment conducted in a real-life matching market, we show that a (small) application cost effectively reduces congestion without sacrificing matching quality.
Behavior seminar
Du 16/11/2017 de 11:00 à 12:00
salle R2-21, campus Jourdan - 48 bd Jourdan 75014 Paris
BRANDTS JORDI (Barcelona Graduate School of Economics )
Centralized vs. Decentralized Management: An Experimental Study
We study the tradeoffs between centralized and decentralized management using a new experimental game, the decentralization game. This game models an organization with two divisions and one central manager. Each division must choose or be assigned a product type. Both divisions benefit from coordinating their product types, but each prefers to coordinate on products that are close to its local tastes. The central manager aims to maximize the sum of division payoffs. Which product type achieves this goal varies with taste shocks that are known to the divisions but not the central manager. Under centralization, the central manager assigns products to divisions after receiving the divisions’ messages about the state of the world (i.e., the taste shock); under decentralization, the divisions choose their own products. Contrary to the theory, overall performance is higher under centralization than under decentralization. Communication between divisions and suggestions from central managers modestly improves performance under decentralization. Nonetheless, centralization remains the best-performing organizational form.