Frictions in a Competitive, Regulated Market

标题：竞争和监管市场中的摩擦。随着优步和Lyft等新技术进入者的崛起，世界各地的出租车市场都处于巨大变革的时代。纽约市等地的出租车市场在某些重要方面类似于竞争市场；特别是大量独立参与者的存在，这些参与者并不具备市场支配力。然而，它们也存在障碍或摩擦，阻碍了这种竞争的顺利运作。其中一些摩擦是监管方面的，比如进入、定价和所有权的限制。有些是产品交付方式的特点。例如，出租车司机对潜在乘客的位置了解有限，导致了浪费的搜索或等待，这就产生了一个重要的“匹配摩擦”。这个项目开发和评估了一个试图衡量这些摩擦的模型，并讨论了可能减少这些扭曲的政策。它还考虑了优步等进入者的影响。PIs之所以研究纽约市，是因为可以获得极其详细的数据，也因为这个市场有一些有趣的特征。该项目的第一部分估计了一个具有匹配摩擦的纽约市出租车市场的动态一般均衡模型。该项目的第二部分探讨了更加依赖于数据中详细的空间信息的问题。PI讨论司机如何协调他们的决策，以及他们如何很好地使用可用的信息。他们提出了一种新的方法，以应对直接观察需求潜力的限制。因为出租车搜索乘客的时间是被观察到的，所以PI可以利用搜索是一个明确定义的空间过程的事实来恢复等待的顾客的数量。在供应方面，潜在司机每天都会做出进入决定，每小时会做出停车决定。然后，必须对这些收入进行汇总，以与均衡收益保持一致。作为市场竞争均衡的一部分，等待时间和出租车数量是由内生共同决定的。PI使用来自纽约市的数据，其中包含从2010年到2013年每一次出租车出行的信息。他们提出了一些反事实的措施，放松对进入和所有权的限制。他们还提出了一个改变匹配技术的反事实：一个集中式调度员，将空出租车发送给最近的乘客。他们考察了效率的提高以及乘客和出租车等待时间的减少，这是调度员控制的市场份额的函数。这与思考优步等公司的影响是相关的。在该项目的第二部分，投资促进机构建议系统地探讨不同类型的集中匹配程序下的结果如何变化，以更好地了解各种摩擦的作用和重要性。他们还可以获得关于出租车之间(均衡)协调程度的信息。可能存在多个帕累托排序的均衡，具有不同程度的市场细分。该项目旨在评估这种市场细分的效果。除了对纽约市出租车市场的明显政策影响外，该项目还有许多潜在的影响。它可以让研究人员量化一些现有法规的扭曲，例如票价结构的僵化。随着最近技术的进步，许多市场现在面临着替代组织模式的可能性。这种结构方法使私人投资机构能够探索这种替代组织的潜力。这些都与其他存在空间摩擦和需要快速提供服务的市场相关。

Title: Frictions in a Competitive, Regulated Market.This is a time of great transformation in taxi markets everywhere, with the rise of new technology entrants such as Uber and Lyft. Taxi markets in places like New York City resemble a competitive market in some important ways; particularly the presence of a large number of independent participants who do not have market power. However, they also feature impediments, or frictions, to the smooth functioning of such competition. Some of these frictions are regulatory, such as the limits to entry, pricing, and ownership. Some are features of the way the products are delivered. For instance, there is an important "matching friction" that derives from the fact that taxi drivers have limited knowledge about the location of potential passengers, leading to wasteful search or waiting. This project develops and estimates a model that attempts to measure these frictions and discusses policies that may reduce these distortions. It also considers the effects of entrants such as Uber. The PIs study New York City because of the availability of extremely detailed data and because of some interesting features of that market. The first part of the project estimates a dynamic general equilibrium model with matching frictions of the New York City taxi-cab market. The second part of the project explores questions that rely more intensively on the detailed spatial information in the data. The PIs discuss how well drivers coordinate their decisions, and how well they use available information. They propose a novel approach to deal with limitations on the potential to directly observe demand. Because the amount of time a taxi searches for passengers is observed, the PIs can recover the number of waiting customers by using the fact that search is a well-defined spatial process. On the supply-side, potential drivers make a daily entry decision and an hourly stopping decision. These must then be aggregated to be consistent with equilibrium earnings. The wait time and the number of taxis are jointly determined endogenously as part of the competitive equilibrium in this market. The PIs use data from New York City with information on every single cab trip from 2010 to 2013. They propose counterfactuals relaxing entry and ownership restrictions. They also propose a counterfactual that changes the matching technology: a centralized dispatcher who sends empty cabs to the closest passenger. They examine gains in efficiency with reductions in wait times for both passengers and taxis as a function of the fraction of the market controlled by the dispatcher. This is relevant for thinking about the effect of firms such as Uber. In the second part of the project the PIs propose to systematically explore how outcomes change under different types of centralized matching procedures to better understand the role and importance of various frictions. They can also obtain information about the (equilibrium) degree of coordination among taxis. There can be multiple Pareto-ranked equilibria with different degrees of market segmentation. The project aims to evaluate the effects of this market segmentation. This project has many potential impacts beyond the obvious policy implications for the NYC taxi market. It can allow researchers to quantify the distortions of some existing regulations, such as the inflexibility of the fare structures. With recent advances in technology, many markets are now faced with the possibility of alternative modes of organization. This structural approach allows the PIs to explore the potential for such alternative organization. These have relevance for other markets with spatial frictions and a need for rapid delivery of service.