An empirical analysis of trends in female labour force participation and the gender wage gap in South Africa

Abstract

In this analysis, we constructed cohort data from repeated cross-sections using the Post-Apartheid Labour Market Series (PALMS) and examined trends in labour market outcomes in the South African labour market. Labour force participation trends show a substantial and persistent gender gap over the life cycle. The gender gap in employment is wide at young ages but seems to decline over the life cycle confirming that younger women are more vulnerable to high unemployment. Earnings trajectories of more recent cohorts of men and women are above those of older cohorts showing that younger cohorts have experienced gains in earnings over time. The gender wage gap is smaller for the youngest cohorts an indication of generational effects. This result suggests a long-term narrowing of the gender wage gap as male and female labour market characteristics become similar. The significance of age is evident as the gender wage gap increases over the life cycle indicating a discontinuous labour force participation for women. Interventions to reduce this discontinuity require among other things alleviating the disproportionate burden of care work shouldered by women to enable them to commit more time to the labour market. A cohort analysis of the occupational distribution by gender shows that more recent cohorts of women are in better occupations than the generations of women before them. This is attributable to improved human capital and better labour market opportunities for younger cohorts. A further narrowing of the gender wage gap requires interventions that will ensure a better distribution of women in male dominated occupations. We note however, that African men and women seem to be dropping out of the labour market at an early age which implies that other factors aside from motherhood or family responsibilities (possibly health issues) are causing this decline in participation.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 38 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.