The Distributional Effects of Prices

该项目提出了两个问题：(1)价格变化对不平等和生活水平的影响是什么？(2)政策引发的价格效应在多大程度上改变了这些政策的成本效益分析？尽管进行了广泛的研究，但我们目前缺乏详细的数据以及各种经验和理论工具来适当地回答这些问题。这些问题至关重要，因为大家都知道，收入分布不同的个人会购买不同种类的商品和服务。因此，随着时间的推移，价格或产品可用性的变化可能对不平等产生重要影响。同样，各国之间的价格差异也会对各国的生活水平产生深远的影响。调查这些问题的少数研究使用了“宏观”数据（在高水平的聚合），但我在之前的工作（Jaravel 2017）中已经表明，使用“微观”数据（即非常分解的数据，在产品层面）来准确回答这些问题至关重要。我们知道，政策可能会产生很大的价格影响（见Jaravel 2018年关于食品券的价格影响）。例如，提高进口关税可能会导致国内消费者的价格上涨（我已经开始在正在进行的工作中进行调查：Borusyak和Jaravel 2017和Jaravel和Sager 2018）。但我们对这种影响可能有多大还没有很好的了解。同样，其他重要的政策，如收入再分配计划或货币政策，可能对价格产生重大影响，这一点目前还没有得到很好的理解。要回答这个项目中提出的两个基本问题，主要有两个挑战。首先，正确衡量价格随时间和国家的变化并不容易，因为可用的商品和服务总是在变化，需要详细的微观数据。其次，由于存在反馈循环，理解政策对价格的影响具有挑战性。例如，如果一个给定的政策使某一特定群体的个人变得富裕，他们可能会改变他们的消费模式，开始购买一套不同的商品或服务，这可能会影响其他代理人的收入，从而改变他们的消费模式，等等。在这个项目中，我建议分两步进行，依次解决这两个主要挑战，以提高我们对价格变化的影响及其对主要政策的影响的理解。该项目的第一部分旨在解决关于测量“质量调整”价格变化的文献中的三个基本限制（基于Jaravel 2017）：(i)各国扫描仪数据的可用性有限；（ii）有限使用享乐回归；（三）对房价变化对福利影响的理解有限。利用新的模型和新的实证工具，该项目的第二部分旨在重新阐明三种重要类型的政策对福利的影响，考虑到它们的价格效应：(i)最优收入和商品税；（ii）贸易政策（以Borusyak and Jaravel 2017和Jaravel and Sager 2018为基础）；（三）货币政策。这个项目的各个部分构成了一个有凝聚力的整体。采取多方面的方法是在理解价格影响及其政策含义方面取得重大进展的唯一途径。这个项目有很大的潜在影响。特别是，它可能会改变国家政府机构公布的通胀统计数据的类型，以及国际组织公布的各国生活水平统计数据的类型。为了确保其他研究人员、政策制定者、智库以及公众能够轻松获取该项目的新数据和新发现，结果和数据将在一个专门的、用户友好的网站上在线提供。

This project asks two questions: (1) What are the implications of prices changes for inequality and standards of living? (2) To what extent do the price effects induced by policies alter the cost-benefit analysis of these policies? Despite extensive research, we currently lack detailed data as well as various empirical and theoretical tools to appropriately answer these questions. These questions are fundamental because it is well-known that individuals across the income distribution purchase different baskets of goods and services. Therefore, changes in prices or product availability over time can potentially have an important impact on inequality. Likewise, differences in prices across countries can have a profound impact on standards of living across countries. The few studies that have investigated these questions have used "macro" data (at a high level of aggregation), but I have shown in previous work (Jaravel 2017) that it is crucial to use "micro" data (i.e. very disaggregated data, at the product level) to accurately answer these questions.We know that policies may have large price effects (see Jaravel 2018 on the price effects of food stamps). For instance, increasing import tariffs is likely to result in higher prices for domestic consumers (which I have started investigating in ongoing work: Borusyak and Jaravel 2017 and Jaravel and Sager 2018). But we do not have a good understanding of how large this effect might be. Likewise, other important policies like income redistribution schemes or monetary policy could have significant effects on prices, which are not well understood currently. There are two main challenges to answer the two fundamental questions asked in this project. First, it is not easy to properly measure how prices change over time and across countries, because the set of available goods and services is always changing and detailed micro data is required. Second, it is challenging to understand the impact of policies on prices because of feedback loops. For instance, if a given policy makes a particular group of individuals richer, they might change their consumption patterns and start buying a different set of goods or services, which may have an impact on the income of other agents, who in turn will change their consumption patterns, etc.In this project, I propose to proceed in two steps, tackling each of these two major challenges in turn to advance our understanding of the effects of price changes and of their implications for major policies. The first part of the project aims at addressing three fundamental limitations in the literature on the measurement of "quality-adjusted" price changes (building on Jaravel 2017): (i) limited availability of scanner data across countries; (ii) limited use of hedonic regressions; and (iii) limited understanding of the welfare impact of house prices changes. Using new models and new empirical tools, the second part of the project aims at shedding new light on the welfare impact of three important types of policies, given their price effects: (i) optimal income and commodity taxation; (ii) trade policy (building on Borusyak and Jaravel 2017 and Jaravel and Sager 2018); and (iii) monetary policy.The various parts of this project constitute a cohesive whole. Taking a multi-faceted approach is the only way of making significant progress on understanding the effects of prices and their policy implications.This project has a strong potential for impact. In particular, it could change the type of inflation statistics published by national government agencies, as well as the type of standards-of-living statistics across countries published by international organisations. To ensure that the new data and new findings from the project are easily accessible by other researchers, policymakers, think tanks, as well as by the general public, the results and data will be made available online on a dedicated, user-friendly website.