Private Disclosures in Competing Mechanisms: Theory and Applications

本研究研究了广泛的一类经济情况下，多个设计师竞争，提供复杂的机制，以吸引和服务于多个私人知情的代理。例子包括卖家设计拍卖以吸引投标人，平台设计支付方案以吸引用户和开发商，地方政府设计税收和补贴以吸引公司到他们的城市，证券交易所设计交易协议和支付方案以吸引买家和卖家到他们的场所。该研究探讨了一个新的维度在这些环境中的机制设计，即可能性，告知有关代理人不对称的机制的运作。例如，拍卖人可能会向一些出价人透露他们的保留价，而对其他人保密，这是一些供应商近年来开始采用的做法，但其优点尚未得到调查。或者，制造商可以通过不对称地告知普通零售商他们的生产如何对市场条件做出反应来提高他们的利润，从而接近合谋的结果。该研究开发了一种新的方法来研究这些市场，并将其用于实证和规范分析。在积极的方面，它调查了在合同分散和使用复杂机制的市场中竞争的局限性。在规范方面，分析揭示了现有法规的优点，如强制性披露法律，排他性，透明度和“满足竞争”的要求，并调查了新的政策干预的福利影响。预计该方法和结果将引起经济学家、计算机科学家和监管机构的兴趣。私人披露（即在代理人在机制中采取行动之前，向代理人不对称地披露有关机制要素的信息）是机制设计理论中的一个新维度。它对均衡结果有着深远的影响。例如，当设计者被限制在具有任意丰富的消息空间但没有私人披露的标准机制时，可以在均衡中维持的分配和收益可能不是鲁棒的，也就是说，当设计者可以参与私人披露时，可能是不可持续的。此外，在允许私下披露的情况下，标准机制无法维持平衡的分配和回报可以维持。这些结果要求一种新的方法来研究竞争机制，这是根据该补助金。这项研究涉及两个方面。在理论方面，分析开发了一种新的方法来描述这些游戏中的均衡分配。它确定了一类易于处理的机制，这是“典型的”，因为它支持所有的均衡结果。在应用方面，该研究显示了私人披露如何在许多感兴趣的市场中得到具体应用，以及监管机构在考虑私人披露时应如何重新考虑这些市场中的某些政策干预。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

This research studies a broad class of economic situations in which multiple designers compete by offering sophisticated mechanisms to attract and serve multiple privately-informed agents. Examples include sellers designing auctions to attract bidders, platforms designing payment schemes to attract users and developers, local governments designing taxes and subsidies to attract companies to their municipalities, and stock exchanges designing trading protocols and payment schemes to attract buyers and sellers to their venues. The research explores a novel dimension in the design of mechanisms in these environments, namely the possibility to inform the relevant agents asymmetrically about the functioning of a mechanism. For example, auctioneers may disclose their reserve price to some bidders while keeping it secret from others, a practice that some vendors have started following in recent years but whose merits have not been investigated yet. Alternatively, manufacturers may raise their profits approaching the collusive outcome by informing common retailers asymmetrically of how their production responds to market conditions. The research develops a new methodology to study these markets and uses it for positive and normative analysis. On the positive front, it investigates the limits to competition in markets where contracting is decentralized and sophisticated mechanisms are used. On the normative front, the analysis sheds light on the merits of existing regulations such as mandatory disclosure laws, exclusivity, transparency, and “meet the competition” requirements, and investigates the welfare implications of novel policy interventions. The methodology and results are expected to be of interest to economists, computer scientists, and regulatory authorities. Private disclosures (that is, information about elements of a mechanism disclosed asymmetrically to the agents before the latter act in a mechanism) is a new dimension in the theory of mechanism design. It has profound effects on equilibrium outcomes. For example, allocations and payoffs that can be sustained in equilibrium when the designers are restricted to standard mechanisms with arbitrarily rich message spaces but no private disclosures may not be robust, that is, may not be sustainable when designers can engage in private disclosures. Furthermore, allocations and payoffs that cannot be sustained in equilibrium with standard mechanisms can be sustained when private disclosures are allowed. These results call for a novel approach to the study of competition in mechanisms that is developed under this grant. The research involves two fronts. On the theoretical front, the analysis develops a new approach to the characterization of equilibrium allocations in these games. It identifies a class of tractable mechanisms that is “canonical” in that it supports all equilibrium outcomes. On the applied front, the research shows how private disclosures can be concretely used in many markets of interest and how regulatory authorities should reconsider certain policy interventions in such markets when accounting for private disclosures.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.