A Generation in Waiting—Unlocking the Employment Potential of Jordan’s Population¹

This note documents Jordan’s labor market outcomes, quantifies the daunting employment challenge ahead, and projects outcomes under alternative reform scenarios by drawing on the international experience. Its results indicate that, by pursuing a wide range of reforms in labor and product markets, Jordan can unlock the employment potential of its population.

1. Unemployment is a chronic problem. It has been persistently high, averaging 13.7 percent in the last decade and registering 12.6 percent in 2013. Unemployment in Jordan is largely a youth phenomenon. Young people, ages 15 to 24, account for almost half of the unemployed, and the youth unemployment rate, at 31 percent, is among the highest in the world. In contrast to most of the world, unemployment rates in Jordan tend to be highest among the educated, registering 17.8 percent among those with college degrees, and for females, exceeding 25 percent (Figure 1).

Unemployment by Gender and Educational Level, 2013

(In percent)

Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

Sources: Jordanian authorities; and IMF staff estimates.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Unemployment by Gender and Educational Level, 2013

(In percent)

Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

Sources: Jordanian authorities; and IMF staff estimates.

• Download Figure
• Download figure as PowerPoint slide

Unemployment by Gender and Educational Level, 2013

(In percent)

Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

Sources: Jordanian authorities; and IMF staff estimates.

• Download Figure
• Download figure as PowerPoint slide

2. Employment outcomes are poor (Figure 2). At 37.1 percent, the labor force participation rate in Jordan is lower than the MENA average of 49.0 percent—and at 13.2 percent, the female labor force participation is less than a quarter of the male participation. Such a low rate, together with high unemployment, has resulted in a very low ratio of employment to working-age population—only 32.4 percent of working-age people are actually employed, a rate that is lower than the MENA average of 43.4 percent and falls way short of the world average of about 60 percent.

Labor Market Outcomes: Jordan and the World, 2013

(In percent)

Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

Sources: For regional averages and Tunisia’s youth labor market outcomes: International Labor Organization, Global Employment Trends 2014. For individual countries, national authorities (Egypt, Central Agency for Public Mobilization and Statistics; Jordan, Department of Statistics; Morroco, Haut-Commissariat Au Plan; Tunisia, Institut National de la Statistique). Notes: DE & EU: Developed Economies & European Union; CSEE& CIS: Central & South-Eastern Europe (non-EU) & Commonwealth of Independent States; EA: East Asia; SEAP: South-East Asia & the Pacific; SA: South Asia; LAC: Latin America & the Caribbean; MENA: Middle East and North Africa; SSA: Sub-Saharan Africa. For Egypt, data are for 2012.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Labor Market Outcomes: Jordan and the World, 2013

(In percent)

Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

Sources: For regional averages and Tunisia’s youth labor market outcomes: International Labor Organization, Global Employment Trends 2014. For individual countries, national authorities (Egypt, Central Agency for Public Mobilization and Statistics; Jordan, Department of Statistics; Morroco, Haut-Commissariat Au Plan; Tunisia, Institut National de la Statistique). Notes: DE & EU: Developed Economies & European Union; CSEE& CIS: Central & South-Eastern Europe (non-EU) & Commonwealth of Independent States; EA: East Asia; SEAP: South-East Asia & the Pacific; SA: South Asia; LAC: Latin America & the Caribbean; MENA: Middle East and North Africa; SSA: Sub-Saharan Africa. For Egypt, data are for 2012.

• Download Figure
• Download figure as PowerPoint slide

Labor Market Outcomes: Jordan and the World, 2013

(In percent)

Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

Sources: For regional averages and Tunisia’s youth labor market outcomes: International Labor Organization, Global Employment Trends 2014. For individual countries, national authorities (Egypt, Central Agency for Public Mobilization and Statistics; Jordan, Department of Statistics; Morroco, Haut-Commissariat Au Plan; Tunisia, Institut National de la Statistique). Notes: DE & EU: Developed Economies & European Union; CSEE& CIS: Central & South-Eastern Europe (non-EU) & Commonwealth of Independent States; EA: East Asia; SEAP: South-East Asia & the Pacific; SA: South Asia; LAC: Latin America & the Caribbean; MENA: Middle East and North Africa; SSA: Sub-Saharan Africa. For Egypt, data are for 2012.

• Download Figure
• Download figure as PowerPoint slide

3. Looking forward, the employment challenge is daunting. To absorb the new entrants to the labor force, and abstracting from any future potential labor market pressures exerted by the Syrian refugees,² Jordan will need to increase employment by an estimated 403 thousand full-time positions over 2013–2020. Under the assumption of a constant employment elasticity set at its empirically estimated historical value of about 0.5 (Table 1),³ staff estimates that creating that many jobs would require an average annual real GDP growth of 6.1 percent. Current growth forecasts would only generate 275 thousand jobs.

Table 1.

Estimating Jordan’s Employment Elasticity¹

Source: IMF staff estimates based on raw data from national authorities and the International Labor Organization.

^1/A vector autoregression involving log employment, log real GDP, and their lags is first estimated, and then Johansen’s cointegration test is applied. The table reports the trace and max test statistics that test the null hypothesis of no cointegration, their small sample variants (labeled T-nm), as well as their p-values. The null hypothesis of no cointegration is strongly rejected. The estimated cointegrating vector puts the employment elasticity at about 0.55, which is highly significant. Furthermore, the evidence suggests that real GDP is weakly exogenous with respect to the parameters of the cointegrating vector. We have experimented with variants of the model that include a deterministic time trend to account for potentially omitted variables, but it was not found to be significant. The residuals of the model are white noise and its estimated coefficients are highly stable over time.

Table 1.

Estimating Jordan’s Employment Elasticity¹

Testing for cointegration between the logarithms of employment and real GDP
Rank	Trace test	P-value	Max test	P-value	Trace test(T-nm)	P-value	Max test (T-nm)	P-value
0	19.13	1 0.00	0.012*	19.13	0.007**	17.39	0.024*	17.39	0.014*
2	0.00	0.983	0.00	0.983	0.00	0.984	0.00	0.984
Cointegration analysis, 1991-2012
Variables			Cointegrating vector		Likelihood ratio test for		P-value
Log Employment			1		…		…
Log RGDP			0.55		Chi^2(1) = 10.023		0.0015**
Testing for Weak Exogeneity
Variables			Speed of adjustment coefficients Likelihood ratio test for significance				P-value
Log Employment			−0.28		Chi^2(1) = 18.441		0.0000**
Log RGDP			−0.05		Chi^2(1) = 0.58287		0.4452

Source: IMF staff estimates based on raw data from national authorities and the International Labor Organization.

^1/A vector autoregression involving log employment, log real GDP, and their lags is first estimated, and then Johansen’s cointegration test is applied. The table reports the trace and max test statistics that test the null hypothesis of no cointegration, their small sample variants (labeled T-nm), as well as their p-values. The null hypothesis of no cointegration is strongly rejected. The estimated cointegrating vector puts the employment elasticity at about 0.55, which is highly significant. Furthermore, the evidence suggests that real GDP is weakly exogenous with respect to the parameters of the cointegrating vector. We have experimented with variants of the model that include a deterministic time trend to account for potentially omitted variables, but it was not found to be significant. The residuals of the model are white noise and its estimated coefficients are highly stable over time.

View Table

Table 1.

Estimating Jordan’s Employment Elasticity¹

Testing for cointegration between the logarithms of employment and real GDP
Rank	Trace test	P-value	Max test	P-value	Trace test(T-nm)	P-value	Max test (T-nm)	P-value
0	19.13	1 0.00	0.012*	19.13	0.007**	17.39	0.024*	17.39	0.014*
2	0.00	0.983	0.00	0.983	0.00	0.984	0.00	0.984
Cointegration analysis, 1991-2012
Variables			Cointegrating vector		Likelihood ratio test for		P-value
Log Employment			1		…		…
Log RGDP			0.55		Chi^2(1) = 10.023		0.0015**
Testing for Weak Exogeneity
Variables			Speed of adjustment coefficients Likelihood ratio test for significance				P-value
Log Employment			−0.28		Chi^2(1) = 18.441		0.0000**
Log RGDP			−0.05		Chi^2(1) = 0.58287		0.4452

Source: IMF staff estimates based on raw data from national authorities and the International Labor Organization.

^1/A vector autoregression involving log employment, log real GDP, and their lags is first estimated, and then Johansen’s cointegration test is applied. The table reports the trace and max test statistics that test the null hypothesis of no cointegration, their small sample variants (labeled T-nm), as well as their p-values. The null hypothesis of no cointegration is strongly rejected. The estimated cointegrating vector puts the employment elasticity at about 0.55, which is highly significant. Furthermore, the evidence suggests that real GDP is weakly exogenous with respect to the parameters of the cointegrating vector. We have experimented with variants of the model that include a deterministic time trend to account for potentially omitted variables, but it was not found to be significant. The residuals of the model are white noise and its estimated coefficients are highly stable over time.

4. What is needed is a combination of permanently higher economic growth and reforms to improve the responsiveness of the labor market to growth. The fact that unemployment has remained high for so long suggests that the problem is largely structural and will not be resolved by a cyclical increase in output. Indeed, over at least the past two decades, while the correlation between unemployment and the output gap has been statistically significant, the implied unemployment rate in the presence of a zero output gap—that is, the structural component of the unemployment rate—has been large, roughly matching the unemployment rate’s historical average (Table 2). Moreover, the concentration of unemployment among the youth and educated suggests that any solution will need to involve greater labor market flexibility and educational reforms.

Table 2.

Unemployment: Structural Component and Responsiveness to the Output Gap

Source: IMF staff estimates based on raw data from national authorities and the International Labor Organization.

^1/Okun’s coefficient and structural unemployment are based on the following Autoregressive Distributed Lag (ADL) Model involving the unemployment rate (u, in percent) and the output gap (gap): $u_t={\alpha }_{\mathrm{o}}+{\mathrm{\Sigma }}_{i=1}^p{\alpha }_i{u}_{t-i}+{\mathrm{\Sigma }}_{i=0}^p{\beta }_i{\mathit{gap}}_{t-i}+{\varepsilon }_t$, where

Okun’s coefficient = $\frac{{\mathrm{\Sigma }}_{i=0}^p{\beta }_i}{1-{\mathrm{\Sigma }}_{i=1}^p{\alpha }_i}$, and structural unemployment = $\frac{{\alpha }_0}{1-{\mathrm{\Sigma }}_{i=1}^p{\alpha }_i}$.The data are annual. The output gap is measured as the percent deviation of the log of real GDP from its Hodrick-Prescott (HP) trend. The smoothness parameter is set equal to 100. The results are robust to using a smoothness parameter of 12, as suggested by Rand and and Tarp (2002)* for emerging markets and developing economis. The lag length is determined using sequential F-tests, as well as the Schwarz Information Criterion. Statistically, the residuals of the model are well behaved and stationary with no signs of autocorrelation or heteroscedasticity. Moreover, recursive estimation as well as recursive Chow tests show that the estimated coefficients of the model are remarkebly constant over time.* Rand, John and Tarp, Finn, 2002. “Business Cycles in Developing Countries: Are They Different?” World Development, vol. 30(12), pages 2071–2088, December.

Table 2.

Unemployment: Structural Component and Responsiveness to the Output Gap

Country	Okun’s Coefficient1	t-value	t-prob	Structural Unemployment Rate1 (in percent)	t-value	t-prob	Unemploment Rate: Sample Average (in percent)	Sample
Jordan	−0.82	−4.70	0.000	14.55	31.80	0.000	14.03	1987 − 2012

Source: IMF staff estimates based on raw data from national authorities and the International Labor Organization.

^1/Okun’s coefficient and structural unemployment are based on the following Autoregressive Distributed Lag (ADL) Model involving the unemployment rate (u, in percent) and the output gap (gap): $u_t={\alpha }_{\mathrm{o}}+{\mathrm{\Sigma }}_{i=1}^p{\alpha }_i{u}_{t-i}+{\mathrm{\Sigma }}_{i=0}^p{\beta }_i{\mathit{gap}}_{t-i}+{\varepsilon }_t$, where

Okun’s coefficient = $\frac{{\mathrm{\Sigma }}_{i=0}^p{\beta }_i}{1-{\mathrm{\Sigma }}_{i=1}^p{\alpha }_i}$, and structural unemployment = $\frac{{\alpha }_0}{1-{\mathrm{\Sigma }}_{i=1}^p{\alpha }_i}$.The data are annual. The output gap is measured as the percent deviation of the log of real GDP from its Hodrick-Prescott (HP) trend. The smoothness parameter is set equal to 100. The results are robust to using a smoothness parameter of 12, as suggested by Rand and and Tarp (2002)* for emerging markets and developing economis. The lag length is determined using sequential F-tests, as well as the Schwarz Information Criterion. Statistically, the residuals of the model are well behaved and stationary with no signs of autocorrelation or heteroscedasticity. Moreover, recursive estimation as well as recursive Chow tests show that the estimated coefficients of the model are remarkebly constant over time.* Rand, John and Tarp, Finn, 2002. “Business Cycles in Developing Countries: Are They Different?” World Development, vol. 30(12), pages 2071–2088, December.

View Table

Table 2.

Unemployment: Structural Component and Responsiveness to the Output Gap

Country	Okun’s Coefficient1	t-value	t-prob	Structural Unemployment Rate1 (in percent)	t-value	t-prob	Unemploment Rate: Sample Average (in percent)	Sample
Jordan	−0.82	−4.70	0.000	14.55	31.80	0.000	14.03	1987 − 2012

Source: IMF staff estimates based on raw data from national authorities and the International Labor Organization.

^1/Okun’s coefficient and structural unemployment are based on the following Autoregressive Distributed Lag (ADL) Model involving the unemployment rate (u, in percent) and the output gap (gap): $u_t={\alpha }_{\mathrm{o}}+{\mathrm{\Sigma }}_{i=1}^p{\alpha }_i{u}_{t-i}+{\mathrm{\Sigma }}_{i=0}^p{\beta }_i{\mathit{gap}}_{t-i}+{\varepsilon }_t$, where

Okun’s coefficient = $\frac{{\mathrm{\Sigma }}_{i=0}^p{\beta }_i}{1-{\mathrm{\Sigma }}_{i=1}^p{\alpha }_i}$, and structural unemployment = $\frac{{\alpha }_0}{1-{\mathrm{\Sigma }}_{i=1}^p{\alpha }_i}$.The data are annual. The output gap is measured as the percent deviation of the log of real GDP from its Hodrick-Prescott (HP) trend. The smoothness parameter is set equal to 100. The results are robust to using a smoothness parameter of 12, as suggested by Rand and and Tarp (2002)* for emerging markets and developing economis. The lag length is determined using sequential F-tests, as well as the Schwarz Information Criterion. Statistically, the residuals of the model are well behaved and stationary with no signs of autocorrelation or heteroscedasticity. Moreover, recursive estimation as well as recursive Chow tests show that the estimated coefficients of the model are remarkebly constant over time.* Rand, John and Tarp, Finn, 2002. “Business Cycles in Developing Countries: Are They Different?” World Development, vol. 30(12), pages 2071–2088, December.

5. Labor and product market reforms can pay large dividends. Recent empirical work conducted by the IMF shows that countries that undertook large-scale reforms have greatly enhanced their employment elasticity, by up to 0.3.⁴ This means that, if Jordan were to follow suit, it can raise its employment elasticity from 0.5 (under a no further reforms scenario) to about 0.8. At the same time, the evidence shows that reforms have a statistically significant positive impact on potential growth. Staff’s simulations show that if Jordan’s labor and product market policy settings were to converge toward those prevailing in the best performing emerging market economies, then Jordan would raise its potential growth by about 0.7 percentage points. Such a scenario would reduce Jordan’s unemployment rate by 7.6 percentage points over the medium term to 5 percent by 2020 (Figure 3). To put this into perspective, such an unemployment rate is lower than the 2013 world average of 6 percent, and is only slightly higher than today’s average rate of 4.5 percent in East Asia (Figure 2).

Unemployment Forecasts

(In percent)

Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

Source: IMF staff estimates.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Unemployment Forecasts

(In percent)

Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

Source: IMF staff estimates.

• Download Figure
• Download figure as PowerPoint slide

Unemployment Forecasts

(In percent)

Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

Source: IMF staff estimates.

• Download Figure
• Download figure as PowerPoint slide

6. To reap those dividends, Jordan will need to:

• Address skills mismatches through educational reform and training programs. While Jordan has made important strides in providing education to its population, entrepreneurs regularly cite the lack of suitable skills among job applicants as an important constraint to hiring (Figure 4), and unemployment rates are highest among the most educated. This suggests that the education system is failing to produce graduates with marketable job skills. Thus, it will need to focus more on quality, realigning curriculums with private-sector needs. At the same time, to put the currently unemployed into jobs, policymakers can scale up and replicate promising training programs, such as the Job Compact, and those run by the Education for Employment Foundation and Injaz in partnership with the private sector. These policies will increase the supply of skilled labor, and hence, for a given change in GDP growth, skilled employment will increase; to the extent that skilled and unskilled workers are complements, employment of unskilled workers will also be boosted. As a result, such policies would contribute positively to the employment elasticity. They would also result in the augmentation of Jordan’s stock of human capital, thereby raising potential growth.

  

  

  Firms Identifying Labor Skill as a Major Constraint

  (Latest available data)

  Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

  Source: World Bank, Enterprise Survey.Note: JOR: Jordan; EGY: Egypt; MAR: Morocco; MENA: Middle East and North Africa; LAC: Latin America and Caribbean; DEA: Developing and Emerging Asia; CCA: Caucasus and Central Asia; SSA: Sub Saharan Africa; EUAR: European Union and Euro Area; AE: Advanced Economies; EMDC: Emerging Markets and Developing Countries.

  • Download Figure
  • Download figure as PowerPoint slide

  View Full Size

  

  Firms Identifying Labor Skill as a Major Constraint

  (Latest available data)

  Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

  Source: World Bank, Enterprise Survey.Note: JOR: Jordan; EGY: Egypt; MAR: Morocco; MENA: Middle East and North Africa; LAC: Latin America and Caribbean; DEA: Developing and Emerging Asia; CCA: Caucasus and Central Asia; SSA: Sub Saharan Africa; EUAR: European Union and Euro Area; AE: Advanced Economies; EMDC: Emerging Markets and Developing Countries.

  • Download Figure
  • Download figure as PowerPoint slide

  Firms Identifying Labor Skill as a Major Constraint

  (Latest available data)

  Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

  Source: World Bank, Enterprise Survey.Note: JOR: Jordan; EGY: Egypt; MAR: Morocco; MENA: Middle East and North Africa; LAC: Latin America and Caribbean; DEA: Developing and Emerging Asia; CCA: Caucasus and Central Asia; SSA: Sub Saharan Africa; EUAR: European Union and Euro Area; AE: Advanced Economies; EMDC: Emerging Markets and Developing Countries.

  • Download Figure
  • Download figure as PowerPoint slide

• Improve the business climate and the quality of institutions. Jordan ranks poorly relative to elsewhere on key areas such as investor protection, contract enforcement, access to credit, and high costs of business start-ups (Figure 5). As documented by a large body of empirical work, reversing such outcomes would foster investment and competition, thereby boosting output growth and labor demand. At the same time, the empirical evidence also shows that improving the business environment and institutional quality would help move workers out of the informal sector—where they have little protection, usually earn low wages, and engage in low-growth-generating activities—and into the formal sector.⁵

  

  

  

  World Bank Doing Business Indicators: Overall Ranks, 2014 1/

  (Out of 189 countries; lower rank means better outcome)

  Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

  Source: World Bank, Doing Business Database.1/ The World Bank Doing Business indicators should be interpreted with caution due to a limited number of respondents, a limited geographical coverage, and standardized assumptions on business constraints and information availability.

  • Download Figure
  • Download Figure
  • Download figure as PowerPoint slide

  View Full Size

  

  

  World Bank Doing Business Indicators: Overall Ranks, 2014 1/

  (Out of 189 countries; lower rank means better outcome)

  Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

  Source: World Bank, Doing Business Database.1/ The World Bank Doing Business indicators should be interpreted with caution due to a limited number of respondents, a limited geographical coverage, and standardized assumptions on business constraints and information availability.

  • Download Figure
  • Download Figure
  • Download figure as PowerPoint slide

  

  

  

  World Bank Doing Business Indicators: Overall Ranks, 2014 1/

  (Out of 189 countries; lower rank means better outcome)

  Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

  Source: World Bank, Doing Business Database.1/ The World Bank Doing Business indicators should be interpreted with caution due to a limited number of respondents, a limited geographical coverage, and standardized assumptions on business constraints and information availability.

  • Download Figure
  • Download Figure
  • Download figure as PowerPoint slide

  View Full Size

  

  

  World Bank Doing Business Indicators: Overall Ranks, 2014 1/

  (Out of 189 countries; lower rank means better outcome)

  Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

  Source: World Bank, Doing Business Database.1/ The World Bank Doing Business indicators should be interpreted with caution due to a limited number of respondents, a limited geographical coverage, and standardized assumptions on business constraints and information availability.

  • Download Figure
  • Download Figure
  • Download figure as PowerPoint slide

  World Bank Doing Business Indicators: Overall Ranks, 2014 1/

  (Out of 189 countries; lower rank means better outcome)

  Citation: IMF Staff Country Reports 2014, 153; 10.5089/9781498368919.002.A002

  Source: World Bank, Doing Business Database.1/ The World Bank Doing Business indicators should be interpreted with caution due to a limited number of respondents, a limited geographical coverage, and standardized assumptions on business constraints and information availability.

  • Download Figure
  • Download Figure
  • Download figure as PowerPoint slide

• Reform public sector hiring practices and compensation policies. The public sector has been an extraordinarily important source of Jordanian employment—over the past fifteen years, public sector employment shares have ranged from 35 percent to almost 40 percent. Its role has distorted labor market incentives and diverted resources away from a potentially more dynamic private sector. Government hiring practices have typically placed a premium on diplomas over actual skills, influencing educational outcomes and contributing to skill mismatches. At the same time, the comparatively greater job security, higher average wages, and more generous non-wage benefits offered by the public sector have inflated wage expectations among new entrants. In fact, the average public sector wage is 122 percent of the average monthly private sector wage. Except for specialists and managerial level jobs, the public sector pays more than the private sector across all job ladders (see Jordan’s National Employment Strategy 2011–2020). New entrants’ capacity to withstand long periods of unemployment—in anticipation of securing more lucrative opportunities in the public sector—is buoyed by familial support and remittances from abroad.

  Public sector hiring procedures will need to place greater emphasis on skills and competition and less on paper qualifications. Moreover, strengthening the link between compensation and performance and implementing merit-based promotion policies would send the right signals regarding skill formation for young people.

• Reach out to the diaspora. There is a large number of Jordanians working abroad, particularly in the Gulf Cooperation Council. In many cases, these expatriates have not only advanced degrees, but also years of experience in competitive environments. Calling on the skilled diaspora and reversing “brain drain” would increase the supply of skilled labor and thereby help the domestic economy grow and compete globally. Returning members of the diaspora would bring with them their skills, capital, marketing-know how, and innovative ideas. They could open businesses and factories and hire from the pool of the unemployed. Improvements in the business environment, better governance, and better enforcement of intellectual and property rights would help bring back overseas talent. The successful experiences of India and Taiwan in luring back the skilled diaspora suggest that such reforms may need to be complemented with specific policy support⁶

• Improve female’s labor market outcomes. Not only are such outcomes worse than those for Jordanian males, but they are also worse than the average female outcomes in MENA, which, in turn, fare the worst worldwide (see Figure 2). Jordan’s National Employment Strategy suggests that various factors are at play, including employer’s perceptions; limited access to information on job opportunities; the lack of targeted active labor market policies for females; the lack of maternity benefits in the private sector; and the absence of affordable and dependable childcare. Addressing such constraints can go a long way to unlocking the employment potential of Jordanian women.