Collaborative Research: Studies of Contracting and Strategic Behavior

该提案描述了两个项目。第一本《外部性竞争的非合作基础》是莉亚·西格尔和迈克尔·惠斯顿的合作著作。第二部分，“房屋选址的战略预承诺”是由Michael Whinston单独完成的工作。第一个项目研究一个委托人和N个代理人之间的双边合同，在不同代理人与委托人之间的合同存在外部性的情况下。文献中存在两种分析这些设置的方法。一个研究具体的非合作双边谈判博弈。另一个假设N很大，并分析了一种竞争均衡。目前两者都不能完全令人满意。第一种方法存在的问题是，不同的非合作契约游戏会导致困难且通常是多种结果。第二个问题是，所研究的竞争均衡概念并没有像传统的竞争均衡概念那样，在N变大的情况下被证明是合理的。这个项目有两个目标。首先，确定对这些合同设置的预测，这些设置对假设的特定议价过程是稳健的。第二，为这些环境下的竞争平衡建立一个非合作的基础。初步研究表明，我们可以确定两种常用的非合作博弈的稳健均衡预测，并且当竞争均衡存在时，它确实是这些博弈均衡结果的唯一限制。第二个项目是对战略预承诺的实证研究。博弈论涌入经济学的最重要方面之一是，它被用于开发战略预承诺模型。在这样的模型中，一个代理对长期决策变量的选择部分取决于代理的预期，即这个选择将影响后来移动的代理所做的决策。近年来，这些模型已被用于许多经济学领域。然而，尽管它们在理论上得到了广泛的应用，但对战略预承诺理论的实证适用性的研究却少得多。该项目试图在单户住宅建筑商做出选址决定的背景下做到这一点。为什么这是一个有吸引力的领域来寻找这样的影响有几个原因。首先，独栋住宅的选址是永久性的。其次，在影响房主效用的各种因素中，与邻近房屋的距离几乎肯定是一个重要因素。第三，房屋建设决策通常是交错的，因此可以比较当一个地块已经被已经建成的地块完全包围时所做出的位置决策的行为，以及当一个相邻的地块尚未建成时所做出的位置选择。第四，有许多可观察到的因素会影响一块土地上不同地点的可取性，这些因素有助于确定战略效果。最后，本文建立的模型类型可以进一步应用于城市经济学中的问题，例如研究分区管制的影响。

This proposal describes two projects. The first, "Non-cooperative Foundations for Competition with Externalities," is collaborative work between llya Segal and Michael Whinston. The second, "Strategic Precommitment in House Location Choices" is work that will be done solely by Michael Whinston. The first project studies bilateral contracting between one principal and N agents in the presence of externalities among different agents' contracts with the principal. Two approaches to analyzing these settings exist in the literature. One studies specific non-cooperative bilateral bargaining games. The other supposes that N is large and analyzes a type of competitive equilibrium. Neither is currently fully satisfactory. The first approach has the problem that different non-cooperative contracting games lead to difficult, and often multiple, outcomes. The second has the problem that the notions of competitive equilibrium studied have not been justified as a limit of situations in which N gets large, as has the traditional concept of competitive equilibrium. This project has two goals. First, to identify predictions for these contracting settings that are robust to the specific bargaining processes assumed. Second, to develop a non-cooperative foundation for competitive equilibrium in these settings. Preliminary research suggests that one can identify robust equilibrium predictions for two commonly studied non-cooperative games, and that when a competitive equilibrium exists, it is indeed the only limit of equilibrium outcomes of these games. The second project is an empirical investigation of strategic precommitment. Among the most significant aspects of the influx of game theory into economics has been its use to develop models of strategic precommitment. In such models, one agent's choice of a long-lived decision variable is determined in part by the agent's expectation that this choice will affect the decisions made by later-moving agents. Such models have been used in recent years in numerous areas of economics. However, despite their widespread theoretical use, much less attention has focused on examining the empirical applicability of the theory of strategic precommitment. This project attempts to do so in the context of location decisions made by builders of single family houses. There are several reasons why this is an attractive area to look for such effects. First, the location decision that accompanies building a single-family house is extremely permanent. Second, among the various factors that contribute to the utility of the homeowner, the distance from neighboring houses is almost certainly a significant one. Third, house building decisions are typically staggered and so permit a comparison of the behavior of the location decisions made when a parcel is already fully surrounded by already built-upon parcels to the location choices made when one adjacent parcel is still not built-upon. Fourth, there are a number of observable factors that affect the desirability of various locations on a parcel of land that can aid in the identification of strategic effects. Finally, the type of models built here may have further application to issues in urban economics such as the study of the impact of zoning regulation.