The Effect of Labour-Demand Shocks on Women’s Participation in the Labor Force: Evidence from Palestine

Abstract

Two interesting facts emerge from the Palestinian labour market. Educational attainment for women swiftly expanded during the 1999–2011 period, but the labour force participation rate for educated women stagnated––disproportionately so for young educated women. We investigate whether changes in labour demand has contributed to women’s sluggish labour force participation. Our empirical analysis used quarterly labour-force data published by Palestine Central Bureau of Statistics between 2005 and 2011. To explore the causal effect of labour demand shocks, we use Bartik instrumental variable approach. Our analysis provides evidence that changes in the labour demand for educated women, rather than improvement in overall demand, affect their labour force participation. This research has important policy implications regarding the economic empowerment of educated women in Palestine suggesting that improvement in overall demand may not benefit educated women and that boosting demand for this specific cohort is what matters.

Acknowledgements

The authors are also grateful to Luca Tiberti, Jorge Davalos, Guy Lacroix, and PEP peer reviewer for their valuable comments.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. See Verick (2014) for a review of women’s participation in the labour force in the MENA region.

2. Economists often look at supply side effects, highlighting the role of social and cultural barriers for women workers (see for example, Olsen, 2006): lack of childcare facilities and institutional child support (see for example Bick, 2015), spouse’s level of income, expected market wage, and fertility (Klasen & Pieters, 2012).

3. Assaad et al. (2018) explain what they referred to as the ‘MENA paradox,’ which reflects a decrease in the LFP rate for educated women. They suggest that the finding could be explained by negative demand shocks from sectors that typically employed educated women – namely, the public sector.

4. Still, we use data for the first quarter of year 1999 to construct the instrumental variable, see more discussion in Section 2.3.1.

5. Relative employment for young educated women is calculated as number of employed young educated women divided by number of employed young educated men.

6. LFP rate of low-educated women is minimal during the study period, and thus sorting out exogenous variations in the demand for their labour is problematic. As a result, we exclud them from our analysis. We also exclud low-educated men because their labour-market outcomes are largely driven by demand from the Israeli labour market (Mansour, 2010) and they are thus less comparable to educated women; our group of interest.

7. We include in the equatin (1) quarter fixed effects and district linear trend to (partially) account for persistencies in LFP growth. However, we estimated a separate model, to account for the LFP persistent shocks at the district-quarter level, through including the lagged depenedent variable in all estimated models. The results show that the linkages between employment growth and LFP growth are robust. The estimates are available from the authors up on request.

8. We used ISIC-Rev 3.1 two-digit sectors to construct the instrument. These included: 1) sales (50), wholesale (51), and hotel & restaurants (55); 2) retail (52); 3) transportation (60) and telecommunications (64); 4) real estate activities (70), other business (74), computer activities (72), and other community, social, and personal activities and other service activities (91, 92, and 99); 5) public administration and defence (75); 6) education (80); 7) health and social work (85); 8) agriculture and forestry (1 and 2); 9) manufacture of food & beverages (15), manufacture of textiles (17), tanning and dressing of leather (19), and manufacture of furniture (36); 10) manufacture of paper (21), manufacture of other non-metallic mineral products (26), and manufacture of basic metals (27); 11) construction (45); 12) other mining and quarrying (14), manufacture of radio, TV, and communication equipment and apparatus (32), manufacture of medicals (33), recycling (37), electricity, gas, steam and hot water supply (40), and collection, purification, and distribution of water (41).

9. For more discussion on the identification of Bartik shift-share Instrument, see: https://blogs.worldbank.org/impactevaluations/rethinking-identification-under-bartik-shift-share-instrument.

10. The share of service jobs is calculated as the number of jobs created for women in a given group divided by the total employment of that group.

Additional information

Funding

This research work was carried out with financial and scientific support from the Partnership for Economic Policy (PEP) (www.pep-net.org) with funding from the Department for International Development (DFID) of the United Kingdom (or UK Aid), and the Government of Canada through the International Development Research Center (IDRC).