Within-job gender pay inequality in 15 countries

The gender pay gap, a persistent disparity across advanced industrialized nations, has been a subject of extensive research. A prevailing theory posits that the gap arises primarily from the sorting of men and women into different jobs, with men disproportionately concentrated in higher-paying occupations and sectors. While this occupational segregation undoubtedly plays a role, the extent of its contribution compared to within-job pay disparities has been a matter of debate. Germinal research, several decades old and limited geographically, primarily suggested that within-job pay differences for men and women performing identical work for the same employer were minimal. This conclusion shaped policy recommendations, emphasizing job-sorting processes and societal perceptions of value as primary targets for intervention. However, the availability of high-quality linked employer-employee data has been historically limited, hindering a more nuanced understanding of within-job gender pay inequality. This study aims to address this gap by leveraging recent, large-scale datasets from 15 countries to reassess the relative contributions of occupational sorting and within-job pay differences to the overall gender pay gap. The results will have significant implications for policy-making, informing strategies for promoting gender pay equity. The study's updated analysis using recent linked employer-employee data from 15 diverse countries will offer a more current and comprehensive picture of gender pay inequality. This improved understanding is vital for designing effective policies that address this multifaceted issue.

Previous research on gender pay inequality has focused largely on the role of occupational segregation, arguing that differences in pay between men and women stem primarily from their concentration in different occupations, rather than differences in pay for the same work within the same organization. Studies like Petersen & Morgan (1995) highlighted the significant contribution of occupation-establishment sex segregation to the gender wage gap. Other research, such as Meyersson Milgrom et al. (2001), examined equal pay for equal work, comparing Sweden, Norway, and the US. However, these studies often relied on individual-level survey data with limitations in capturing precise job and firm characteristics, making it challenging to accurately isolate within-job pay disparities. Petersen et al. (1997) did explore within-job wage discrimination, but their analysis was limited to Norway and employed older data. The current study builds upon this existing literature by utilizing more recent and comprehensive linked employer-employee data from a wider range of countries, allowing for a more robust and nuanced assessment of the within-job gender pay gap. The study acknowledges contributions from Blau and Kahn (2017) on the overall gender wage gap and England et al. (2020) highlighting stalled progress toward gender equality, and also considers other relevant works such as Reskin and Roos (1990) on job queues and England (1992) on comparable worth.

This study employs linked employer-employee data from 15 countries: Canada, Czechia, Denmark, France, Germany, Hungary, Israel, Japan, the Netherlands, Norway, Slovenia, South Korea, Spain, Sweden, and the United States. The data, sourced from various national statistical agencies and registries, allowed the researchers to identify and compare the earnings of men and women working in the same establishments and occupations. The analyses utilize four sets of ordinary least squares (OLS) regression models. Model 1 adjusts for basic individual-level covariates (age, age-squared, education, and full-time/part-time status). Subsequent models introduce fixed effects for establishment (Model 2), occupation (Model 3), and occupation-establishment unit (job; Model 4). By comparing the results across these models, the researchers quantify the extent to which gender pay differences are attributable to sorting into different establishments, occupations, or jobs, versus within-job pay disparities. The dependent variable is the natural logarithm of earnings, allowing for a direct comparison of relative pay differences. The study carefully addresses potential confounding factors such as part-time employment and variations in data collection methods across countries. Each model is estimated separately by year for each country, allowing analysis of trends in gender pay gaps over time. The natural log of earnings is used as the dependent variable, and the coefficients are interpreted as the relative difference between the average female and male earnings. The study notes that these coefficients more formally represent the difference in relative geometric means for unlogged earnings. Data were analyzed using STATA versions 14–17 and SAS version 9. Robustness checks are performed to validate the findings.

The analysis reveals substantial within-job gender pay gaps across all 15 countries. After adjusting for basic individual characteristics, the overall gender pay gap ranged from 10% (Hungary) to 41% (South Korea). Crucially, even after controlling for establishment, occupation, and job, significant within-job gaps persisted, ranging from 7% (Denmark and France) to 26% (Japan). On average, within-job differences accounted for approximately half of the overall gender pay gap in these countries, ranging from about one-third (Israel) to over nine-tenths (Hungary). The study also finds that sorting into both establishments and occupations plays a significant role in generating gender pay differences, refuting previous research that emphasized the dominance of occupational sorting. Analysis of annual changes in overall and within-job gender pay gaps shows a decline in both over time in most countries but not in three Central and Eastern European countries (Czechia, Hungary, and Slovenia). In Czechia, within-job differences declined while overall differences increased, suggesting increased importance of occupational sorting. In Hungary and Slovenia, increases in both overall and within-job gaps were observed. Notably, no country exhibited a decrease in the overall gender pay gap alongside an increase in the within-job gap, suggesting that occupational sorting processes rarely counteract rising within-job inequality. Table 1 presents detailed gender pay gap estimates for each country across various levels of adjustment (basic, establishment, occupation, and job). Table 2 provides an overview of the key data features used in each country's analysis, including data sources, sample sizes, and key variable definitions. Figure 1 illustrates country-specific trends in overall and within-job gender pay gaps from approximately 2005 to the most recent year of available data.

This study’s findings challenge the long-held assumption that occupational segregation is the primary driver of the gender pay gap. The persistence of substantial within-job pay differences across 15 diverse countries strongly suggests that equal pay policies are needed in conjunction with policies focused on addressing occupational sorting. The fact that within-job differences account for a significant portion of the overall gap, coupled with the observation that sorting is increasingly important, highlights the complexity of the issue. The observed trends, with mostly declining gaps in overall and within-job pay disparities, do not universally apply across all countries, pointing to country-specific factors that influence gender pay inequality. The study's cross-national comparison underscores the importance of addressing both within-job discrimination and broader occupational segregation in efforts to achieve gender pay equity. Further research should explore the specific mechanisms driving within-job pay differences, including the influence of implicit bias, negotiation practices, and other factors influencing pay setting processes within firms. The varying trends observed across countries also necessitate region-specific analysis to improve understanding of contributing factors.

This study provides compelling evidence that within-job gender pay inequality remains a significant contributor to the overall gender pay gap across 15 countries. The findings demonstrate that policies focusing solely on occupational segregation are insufficient to address the issue. Both equal pay legislation and measures targeting hiring, promotion, and other job-sorting practices are vital for achieving gender pay equity. Future research should examine the specific mechanisms underlying within-job disparities and explore context-specific factors driving the observed differences across countries. This study provides a robust foundation for developing more effective policies aimed at achieving fair compensation for all.

The study acknowledges potential limitations related to data harmonization across countries with differing data collection methods, parental leave policies, part-time work prevalence, and occupation and firm structures. While efforts were made to ensure comparability, cross-country comparisons inherently involve varied gender regimes and work organization patterns. However, the authors conducted multiple robustness checks and alternative specifications to demonstrate that the findings are relatively robust to these issues. The available data may not capture all nuances of pay determination processes, particularly subtle forms of discrimination that are hard to detect with readily available data.