Individual-level variables such as gender, education, occupation, and employment type are well-known factors that induce wage gaps in the labor market. This article aims to divide wage gaps into two components—those arising within each firm (the intra-firm wage gap) and those arising between different firms (the inter-firm wage gap)—and measure their respective proportions by individual-level variables, as studies suggest that each wage gap based on an individual-level variable has a unique mix of an intra-firm and inter-firm wage gap. This measurement can help enrich investigation into the generative process of wage gaps and formulate effective labor policies to reduce them. Accordingly, we compared the coefficients of the independent variables of the wage function estimated by a pooled ordinary least squares model and a fixed-effect model, using nationwide employer–employee matching data collected by the Japanese government. We found that wage gaps by gender and employment type mostly consist of intra-firm wage gaps, and those by education and occupation have a larger share of inter-firm wage gaps. The findings suggest that different research strategies are required to investigate the generative process of each wage gap, and that regulations on and interventions in organizational processes are important to reduce wage gaps based on gender and employment type.

中文翻译：

---

个人层面变量引起的工资差距的两个组成部分：企业内部还是企业间？

性别、教育、职业和就业类型等个人层面的变量是众所周知的导致劳动力市场工资差距的因素。本文旨在将工资差距分为两个部分——每个企业内部产生的工资差距（企业内部工资差距）和不同企业之间产生的工资差距（企业间工资差距）——并通过个人层面的变量来衡量它们各自的比例，因为研究表明，基于个人层面变量的每个工资差距都有公司内部和公司间工资差距的独特组合。这种测量有助于丰富对工资差距产生过程的调查，并制定有效的劳动力政策来减少它们。相应地，我们比较了合并普通最小二乘模型和固定效应模型估计的工资函数自变量的系数，使用日本政府收集的全国雇主-雇员匹配数据。我们发现按性别和就业类型划分的工资差距主要由企业内部工资差距组成，而教育和职业的工资差距在企业间工资差距中所占的份额更大。研究结果表明，需要不同的研究策略来调查每个工资差距的产生过程，并且组织过程中的法规和干预对于减少基于性别和就业类型的工资差距很重要。