Do labor markets and the social contract increase female youth unemployment in the GCC countries?

ABSTRACT

In the high-income Gulf Cooperation Council countries, the female youth unemployment rate is nearly double the male youth unemployment rate. The segmented labor market with a relatively inflexible national labor market segment – coupled with a generous social contract – provide job protection to nationals and further the benefits of the social contract. These two factors may shape national youth preferences in favor of protected government jobs and raise youth reservation wages, which increase female youth unemployment. We conjecture that flexible non-segmented labor markets improve female youth unemployment rate while the social contract worsens it. Empirical evidence shows labor market flexibility and the social contract improve female youth unemployment rate, a result robust to changes in model specification and the sample. However, labor market flexibility is essential to the improvement in the female youth unemployment rate, while the social contract is not. This research has an important policy implication for selecting the appropriate tool to address youth unemployment.

KEYWORDS:

• Female youth unemployment
• youth unemployment
• labor markets
• social contract
• Gulf cooperation council countries

Acknowledgements

The author wishes to thank without implication participants at the 1st NYU Abu-Dhabi academic conference, 1st UAE Ministry of Economy Policy Research Seminar, 2021 Global Development Finance Conference, the 37th EBES Conference (Berlin), Jorge Martinez and Mark Gellerson for their comments and suggestions on earlier versions of this paper. All errors are mine.

Disclosure statement

The author declares no financial or personal relationships with other people or organizations that could inappropriately influence this research in any respect.

Notes

1 See Assidmi and Wolgamuth (2017) on the Saudi Arabian experience.

2 The sample of high-income countries include Australia, Austria, Barbados, Belgium, Brunei Darussalam, Canada, Chile, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hong Kong, Hungary, Iceland, Ireland, Israel, Italy, Japan, Korea, Latvia, Lithuania, Luxembourg, Malta, Netherlands, New Zealand, Norway, Panama, Poland, Portugal, Singapore, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Trinidad and Tobago, UK, US, and Uruguay.

3 These numbers are based on author calculations. The GCC statistical center data is available at https://gccstat.org/en/?msclkid=9e9ad3d7b44711ec8af881af76463383

4 See Al-Saidi (2020) on the role that energy subsidies play as a part of the social contract in the GCC countries. See also Al-Sheikh and Erbas (2016) on the high wages and expansive employment policies in Saudi Arabia.

5 Because of missing observations, we interpolate the compensation of government employees using nominal GDP (US$).

6 Loewe et al. (2021) point out that the effectiveness of social contracts depends on (a) the substance or the exchanged deliverables, presumably the benefits to the societal groups, (b) the scope or coverage of actors involved and the geographic range of influence, and (c) the evolution and duration of the social contract over time.

7 The restrictions on female employment in certain sectors in some GCC countries and the higher graduation rates of females compared to males are factors that very likely increases the female youth unemployment rates.

8 In neighboring Iran, where society culture and norms are close, discrimination in the workplace is rampant (Hedayat et al., 2013).

9 The nationals of the Maldives and the GCC countries face similar labor issues. For an interesting study, see Salvini et al. (2016).

10 The African economies are characterized by large informal sector. He explains youth unemployment rate in terms of supply factors (the share of youth in total population and gross enrolment rate), demand factors (real GDP growth rate and the share of agriculture and manufacturing in GDP) and labour market variables (ratio of employment to population and the vulnerable employment rate).

11 Marelli and Vakulenko (2016), Mendolia and Walker (2015) and Mursa et al. (2018) highlight the importance of personal and family characteristics to youth employment.

12 Among the dependent variables was youth unemployment rate, which was explained in terms of the lagged dependent variable, a vector of control variables, a vector of labor market flexibility variables, and a vector of fiscal consolidation variables. The control and labor market flexibility variables were similar to Bernal-Verdugo et al. (2012). They used fixed effects and Arellano-Bond GMM estimator.

13 They used a two-step system generalized method of moments (GMM) estimation methodology to account for the lagged dependent variable and potential simultaneity. In doing so, they considered all explanatory variables as endogenous and instrumented them using up to two lags.

14 Because of the paucity of statistical data in some GCC countries, we use the available ILO-modeled female youth unemployment rate, which defines youth in terms of ages 15–24.

15 It should be noted that in the study period expatriates had to leave the country once the labor contract ends or is terminated. Accordingly, this practice reduces the number of unemployed and the size of the labor force, and likely the unemployment rate. With the unemployed (expatriates) forced to leave the GCC countries, the unemployment rate may underestimate the actual unemployment rate had the expatriates been rather allowed to stay in the country.

16 Exports have a similar impact on unemployment in developed economies, economies with low industry ratio, and economies with high service ratios.

17 The memo items in the table show the national estimates. The estimates for Saudi Arabia and Qatar are close to the International Labor Organization (ILO) estimates. For other GCC countries, insufficient observations are available.

18 The influence on FYUR would be the product of the estimated coefficient and the average value of the variable. In the case of linking pay to productivity, the average value of linking pay to productivity (log) is 1.51. Accordingly, the influence on FYUR ranges from 1.6 to 2 percentage points, depending on whether we multiply the average value by −108.112 (Table 6) or −132.53 (Table 7).

19 The average values of each of these indicators (log) are 1.45, 1.60, 1.74, and 1.51, respectively.

20 By ‘unmatching’, we mean jobs that are not of equal levels of salary or protection.

21 This result holds whether we measure the social contract using compensation of government employees (percent of government expense) or oil rents (percent of GDP).