Doctoral Dissertation Research in Economics: The Impact of Product Market Competition on the Labor Market

该奖项将资助一篇博士论文，该论文使用独特的数据来衡量高市场集中度对工人的影响，包括更好地理解一家公司面临的竞争之间的联系，因为它销售的产品和它面临的竞争，它寻求雇用的工人。经济理论预测，当有许多雇主提供工作时，工人将获得更高的工资。然而，在一些地方，只有几个雇主。在这种集中的劳动力市场中，雇主具有优势，因为工人缺乏可行的替代工作。此外，如果一家公司在其销售的产品市场上占据主导地位，那么它也可能在劳动力市场上占据主导地位。这样的公司将对其员工和消费者都发挥巨大的作用。该项目将系统地调查集中度如何影响工资、就业持续时间和利润传递。该项目的结果可能有助于决策者了解旨在降低市场集中度的政策的成本和收益。该项目基于2008年至2020年的三个管理数据集。第一个是准宇宙的就业咒语逐年增加；研究团队将以此为基础系统构建就业和工资账单集中度指标。出于保密原因，第一个数据集缺乏长期跟踪员工的能力。第二个数据集是一个雇主-雇员面板，包含工资、工作时间、行业、职业和一些人口统计信息。这个数据集使我们能够多年来跟踪工人和公司。最后，第三个数据集包含公司层面的资产负债表信息（销售、利润、库存等）。这些数据集的独特之处在于可以使用唯一的公司标识符合并它们。该项目建立了劳动力市场集中度指数，并将其与我们的面板数据和每家公司的资产负债表合并。结果是一个新的面板，我们可以看到工资是如何随着产品和劳动力市场的集中而变化的。这个分析代表了项目的核心。第一步将是正确定义劳动力市场。我们将依赖传统的概念，如通勤区域的职业，以及基于员工跨公司转换的聚类方法（在随机块建模中应用新开发的技术）。然后，我们将运行劳动力回归来记录工资（水平和增长）、集中度、利润和销售之间的模式。我们想要证明，在更集中的市场中，工人的状况是更好还是更差。如上所述，在更集中的市场中，企业可能有更大的议价能力，但它们也应该是生产率更高的企业，因此支付更高的工资。然而，这些回归并不能告诉我们市场集中度对工资的因果关系。为了缓解这一点，该项目的下一步将是提出一个搜索和匹配模型，其中离散数量的公司竞争工人，并使用我们的实证结果来校准它。然后，我们将能够对集中度对工人福利的影响进行反事实分析。该奖项反映了美国国家科学基金会的法定使命，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 award will fund a doctoral dissertation that uses unique data to measure the effects of high market concentration on workers, including work to better understand the links between the competition a firm faces for the products it sells and the competition it faces for workers it seeks to employ. Economic theory predicts that workers will earn higher wages when there are many employers offering jobs. However, in some local areas there are only a few employers. In this kind of concentrated labor market, employers have an advantage because workers lack viable alternative offers. Furthermore, if a firm has a dominant position in the market for the product/s it sells, it may also be dominant in the labor market. This kind of firm would then have an outsized role on both its workers and its consumers. The project will systematically investigate how concentration affects wages, employment spell duration, and profit pass-through. The results of this project may help policymakers understand the costs and benefits of policies designed to reduce market concentration. This project is based on three administrative datasets spanning the years 2008 to 2020. The first is the quasi-universe of employment spells year-by-year; the research team will use this to systematically construct the employment and wage bill concentration indexes. This first dataset lacks the ability to track workers over time for confidentiality reasons. The second dataset is an employer-employee panel with information on wages, duration of spell, industry, occupation and some demographic information. This dataset allows us to follow workers and firms over the years. Finally, the third dataset contains balance sheet information at the firm’s level (sales, profits, inventories, etc.). What makes these datasets unique is the possibility to merge them using a unique firm identifier. The project builds the concentration index by labor market and merges that measure with our panel data and each firm’s balance sheet. The result is a new panel in which we can look at how wages vary depending on concentrations in the product and labor market. This analysis represents the core of the project. The first step will be to properly define labor markets. We will rely on traditional notions like occupation by commuting zone, in addition to a clustering approach based workers’ transitions across firms (applying newly developed techniques in stochastic block modeling). We will then run labor regressions to document patterns between wages (level and growth), concentration and profits and sales. We want to document whether workers in more concentrated markets are better or worse off. As pointed out above, in more concentrated markets, firms are likely to have more bargaining power, yet they should also be more productive firms, and therefore be paying higher wages. These regressions will not inform us on the causality of market concentration on wages though. To alleviate that point, the next step of the project will be to propose a search-and-matching model in which a discrete number of firms compete for workers and use our empirical results to calibrate it. We will then be able to run counterfactual analyses on the impact of concentration on workers’ welfare.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.