Collaborative Research: Trade Costs, Preferences, and Government Policies: Understanding Market Outcomes in the World Automobile Industry

合作研究：贸易成本、偏好和政府政策：了解世界汽车行业的市场结果

• 批准号: 1459905
• 负责人: Paul Grieco
• 金额: $ 27.76万
• 依托单位: Pennsylvania State Univ University Park
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2015
• 资助国家: 美国
• 起止时间: 2015-05-01 至 2019-04-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1459905&HistoricalAwards=false
• 关键词: Collaborative; Research; Trade; Costs; Preferences

The goal of this project is to quantify the sources of market segmentation, and to understand the implications of trade and investment frictions on the world automotive industry. The framework enables the PIs to assess the importance of trade costs, preferences for domestic brands, and demand heterogeneity. The PIs also quantify the impact of tariff and non-tariff barriers (e.g. arising from differences in fuel prices and environmental regulation) on market outcomes.The project will be the first to develop consistent estimates of automobile demand and marginal costs using data across several continents. This geographical variation allows the PIs to estimate a variety of trade-related costs while accounting for model-level unobserved heterogeneity and home bias. This approach disentangles barriers set by policies from technological barriers such as transport costs. The scope of their analysis will advance the state of knowledge on the interactions between trade costs, preference heterogeneity, and government policies in shaping market outcomes in the automobile industry and in international trade more broadly. The study will evaluate the importance of trade costs versus preference differences in the world trade in automobiles---a major manufacturing industry. This will lead to further insights into understanding the incentives of firms to conduct foreign direct investment and conduct market-specific model development. The dataset with assembly location included will be a valuable resource for analyzing the importance of FDI in the auto industry. The estimates on the importance of tariff and non-tariff barriers in the automotive industry will inform policy makers for the ongoing Trade and Investment Partnership negotiations between the U.S. and the EU. To undertake this research, the PIs first construct a new data set by linking sales data from nine countries on three separate continents, collected by the consulting firm Polk Automotive, with a world census of assembly locations by model---collected by WardsAuto. They link these datasets to yield a model-level data on sales, prices, characteristics, assembly and headquarters location. Second, they use this data to estimate a structural model of international demand for automobiles. Extending techniques pioneered by Berry, Levinsohn, and Pakes (1995), they propose a model that accounts for inter and intra country heterogeneity in tastes for characteristics and income to flexibly estimate elasticities for automobiles in each country. This demand model will measure the importance of "home bias" or the tendency of consumers to prefer their local manufacturers to foreign manufacturers. A key feature of their model is to separate these demand driven explanations from supply side drivers. Third, they use prices and demand elasticities to construct estimates of the marginal cost of supplying each automobile to each market, and use them to estimate the supply side of the model. Going forward, the cost and demand estimates can be used to analyze firms' location and marketing decisions and study the impact of potential policy interventions.

该项目的目标是量化市场细分的来源，并了解贸易和投资摩擦对世界汽车行业的影响。该框架使PI能够评估贸易成本，对国内品牌的偏好和需求异质性的重要性。该项目还量化了关税和非关税壁垒（如燃料价格和环境监管的差异）对市场结果的影响。该项目将是第一个利用几个大洲的数据对汽车需求和边际成本进行一致估计的项目。这种地理差异允许PI估计各种贸易相关成本，同时考虑模型级未观察到的异质性和本土偏差。这一办法将政策设置的壁垒与运输成本等技术壁垒区分开来。他们的分析范围将推进贸易成本，偏好异质性和政府政策之间的相互作用，在汽车行业和更广泛的国际贸易中塑造市场结果的知识状态。这项研究将评估贸易成本的重要性与优惠差异在世界贸易中的汽车-一个主要的制造业。 这将导致进一步深入了解企业进行外国直接投资的动机，并进行针对具体市场的模型开发。包含装配地点的数据集将成为分析汽车行业外国直接投资重要性的宝贵资源。对汽车行业关税和非关税壁垒重要性的估计将为美国和欧盟之间正在进行的贸易和投资伙伴关系谈判的政策制定者提供信息。为了进行这项研究，PI首先通过将咨询公司Polk Automotive收集的三大洲九个国家的销售数据与WardsAuto收集的世界车型装配地点普查数据联系起来，构建了一个新的数据集。 他们将这些数据集连接起来，产生关于销售、价格、特性、装配和总部位置的模型级数据。其次，他们使用这些数据来估计国际汽车需求的结构模型。他们扩展了Berry，Levinsohn和Pakes（1995）开创的技术，提出了一个模型，该模型考虑了国家间和国家内对特征和收入的偏好异质性，以灵活地估计每个国家的汽车弹性。这一需求模型将衡量“本土偏好”的重要性，即消费者倾向于选择本国制造商而不是外国制造商。他们的模型的一个关键特征是将这些需求驱动的解释与供应方驱动因素分开。 第三，他们使用价格和需求弹性来估算向每个市场供应每辆汽车的边际成本，并使用它们来估算模型的供应方。 展望未来，成本和需求估计可用于分析公司的选址和营销决策，并研究潜在政策干预的影响。