Consistent Segregation Metrics: Addressing Structural Variations in Global Labor Markets
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917945631834112 |
|---|---|
| author | Kujundzic, Ana Pieters, Janneke |
| author_facet | Kujundzic, Ana Pieters, Janneke |
| contents | The Index of Dissimilarity (ID), widely utilized in economic literature as a measure of segregation, is inadequate for cross-country or time series studies due to its failure to account for structural variations across countries' labor markets or changes over time within a single country's labor market. Building on the works of Karmel and MacLachlan (1988) and Blackburn et al. (1993), we propose a new measure - the standardized ID - that isolates structural differences from true differences in segregation across space or time. A key advantage of our proposed measure lies in its ease of implementation and interpretation, even when working with datasets encompassing a large number of countries or time periods. Moreover, our measure can be consistently applied in the case of lumpy sectors or occupations that account for a large fraction of the workforce. We illustrate the new measure in an analysis of the cross-country relationship between economic development (as measured by GDP per capita) and occupational and sectoral gender segregation. Comparing the crude ID with the standardized ID, we show that the crude ID overestimates the positive correlation between income and segregation, especially between low- and middle-income countries. This suggests that analyses relying on the crude ID risk overestimating the importance of income differentials in explaining cross-country variation in gender segregation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_02763 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Consistent Segregation Metrics: Addressing Structural Variations in Global Labor Markets Kujundzic, Ana Pieters, Janneke General Economics Economics The Index of Dissimilarity (ID), widely utilized in economic literature as a measure of segregation, is inadequate for cross-country or time series studies due to its failure to account for structural variations across countries' labor markets or changes over time within a single country's labor market. Building on the works of Karmel and MacLachlan (1988) and Blackburn et al. (1993), we propose a new measure - the standardized ID - that isolates structural differences from true differences in segregation across space or time. A key advantage of our proposed measure lies in its ease of implementation and interpretation, even when working with datasets encompassing a large number of countries or time periods. Moreover, our measure can be consistently applied in the case of lumpy sectors or occupations that account for a large fraction of the workforce. We illustrate the new measure in an analysis of the cross-country relationship between economic development (as measured by GDP per capita) and occupational and sectoral gender segregation. Comparing the crude ID with the standardized ID, we show that the crude ID overestimates the positive correlation between income and segregation, especially between low- and middle-income countries. This suggests that analyses relying on the crude ID risk overestimating the importance of income differentials in explaining cross-country variation in gender segregation. |
| title | Consistent Segregation Metrics: Addressing Structural Variations in Global Labor Markets |
| topic | General Economics Economics |
| url | https://arxiv.org/abs/2503.02763 |