TL;DR
This paper investigates how changes in the real value of the minimum wage have impacted wage inequality in the U.S. by analyzing both differences between groups (e.g., by education, gender) and within groups (e.g., among workers with similar education). The authors find that minimum wage changes significantly affect wage differentials, particularly for lower-skilled workers, and play a substantial role in explaining observed trends in wage inequality.
What Problem Does This Paper Address?
Globally, there's a growing concern about rising wage inequality. While minimum wage policies are often proposed as a tool to address this, their precise impact on wage structures, especially how they affect different groups of workers and the disparities within those groups, remains a complex question. Previous research has often focused on the aggregate effects or specific aspects of wage inequality. This paper aims to provide a more nuanced understanding by disentangling the effects of minimum wage changes on both between-group and within-group wage differentials, considering how these impacts vary across different worker characteristics and wage levels.
Methods
The authors employ a sophisticated econometric framework that combines several techniques to address the complexities of minimum wage impacts:
- Quantile Regression: Instead of just looking at average wages, they use quantile regression to analyze how the minimum wage affects wages at different points of the wage distribution (e.g., the 10th percentile, the median, the 90th percentile). This is crucial because minimum wage effects are likely to be concentrated at the lower end of the distribution.
- Panel Data Model: They utilize state-level panel data, allowing them to control for state-specific and time-specific factors that might influence wages.
- Censored Quantile Regression: To account for the fact that wages below the minimum wage are effectively "censored" (i.e., they cannot be lower than the minimum), they use a censored quantile regression approach. This helps to correct for potential biases in the estimates.
- Imputation of Missing Wages: The study also addresses the issue of missing wage data by imputing wages for non-employed individuals, using a censored quantile regression model. This is important because changes in employment due to minimum wage hikes could otherwise bias the results.
- Decomposition of Inequality: They develop a method to quantify the contribution of minimum wage changes to both between-group and within-group wage differentials by comparing actual wage differentials with counterfactual scenarios where the minimum wage is held constant.
Datasets
The study uses repeated cross-sectional data from the Current Population Survey (CPS) Merged Outgoing Rotation Group for the period 1979 to 2012. The sample is restricted to workers aged 18-64, including both full-time and part-time workers, but excluding self-employed individuals.
Key Insights and Ideas
- Minimum Wage as a Driver of Inequality Trends: The paper strongly suggests that changes in the real value of the minimum wage are not just a minor factor but a significant contributor to observed patterns in wage inequality, both between and within groups.
- Heterogeneous Impacts: The impact of the minimum wage is not uniform. It disproportionately affects workers at the lower end of the wage distribution and those with less education, experience, or who are female.
- Weakening of Wage-Characteristic Relationships: Increases in the minimum wage tend to weaken the relationship between hourly wages and characteristics like education, experience, and gender. This implies that these factors become less predictive of wage differences when the minimum wage is higher.
- Within-Group Inequality is Heavily Influenced: The study finds that minimum wage changes are particularly important in explaining changes in within-group wage differentials, especially for workers with less experience.
Key Results
- Intercept and Slope Effects: An increase in the minimum wage leads to a rise in the intercept of the wage equation (benefiting lower-skilled workers) but a decrease in the slope coefficients for education, experience, and gender. This means the wage premium associated with these characteristics diminishes.
- Between-Group Inequality: The minimum wage contributes to a reduction in educational, experience, and gender wage differentials, particularly at lower quantiles of the wage distribution. For instance, it reduces the gender wage gap, especially for less-educated and less-experienced workers.
- Within-Group Inequality: The minimum wage significantly reduces within-group wage differentials (e.g., the 90/10 and 50/10 differentials), especially for workers with lower levels of education and experience. The impact is more pronounced for less-educated, less-experienced, and female workers.
Limitations and Future Work
- Exogeneity Assumption: The study assumes the exogeneity of the minimum wage, which is standard in the literature but could be a limitation if unobserved state characteristics drive both minimum wage changes and wage structures. The authors do examine this possibility.
- Missing Wages Imputation: While they impute missing wages, the accuracy of this imputation could still influence the results.
- Future Work: The paper's findings highlight the need to incorporate the influence of the minimum wage when interpreting wage differentials through economic models. Further research could explore the dynamic effects of minimum wage policies over longer periods and across different institutional settings. The authors also suggest exploring the role of interactions between treatment variables (minimum wage) and individual characteristics in more detail.
Summary updated