UK aid’s approach to youth employment in the Middle East and North Africa

Introduction

This literature review informs the Independent Commission for Aid Impact’s (ICAI) assessment of the UK government’s efforts to promote youth employment in the Middle East and North Africa (MENA). It is not intended to be a systematic review but captures the main strands of the literature on the topic, to ground the ICAI review in a broader understanding of ‘what works’ on youth employment in MENA. It focuses, where possible, on international peer-reviewed literature – particularly systematic and meta-analyses – published in the past 15 years by academic experts and international and regional organisations.

We outline key terms relevant to this review in Annex 1. These include our definition of ‘youth’/‘young people’, employment, unemployment, labour force participation, and Not in Employment, Education or Training (NEET) – a common indicator of youth exclusion. The Annex also explores the concept of labour force ‘transitions’, quality of work, sustained employment outcomes and sustained economic development. This last concept encompasses job creation, governance, peace and security and climate resilience (UK Aid Strategy, 2015).

Geographic scope

Where possible, we prioritise evidence relevant to MENA. We caution that definitions of ‘MENA’ vary, with direct consequences for how institutions structure their programmes, research and data. The World Bank (2020) described MENA as comprising the area from Morocco to Pakistan, including Djibouti, Iran, Israel and Malta. The former UK Department for International Development (DFID) describes MENA as “a diverse region” of 19 countries “in which Gulf oil and gas states sit next to poorer countries like Yemen” (DFID, 2013, p. 1). DFID prioritised Yemen, the Occupied Palestinian Territories (OPT), Syria, Jordan and Lebanon in the region. DFID had a lighter footprint in North Africa with work in Libya, Egypt and Tunisia (DFID, 2014).^[1]

MENA is sometimes sub-divided into three categories: the oil-rich ‘Gulf’, the Maghreb, comprising Arabic and French-speaking countries in North Africa and the Sahel, and the Mashreq or Levant, comprising Iraq, Jordan, Lebanon, Syria and Israel/OPT.

While we take the World Bank (2020) definition as a starting point, our research focuses on the following 13 MENA countries of interest, hereafter referred to as ‘the cohort’ or ‘the ICAI Cohort’, grouped by category per Organisation for Economic Co-operation and Development (OECD) classifications (2020). These countries are eligible to receive overseas development assistance (ODA) and include territories at various stages of development, including those in conflict, such as Syria, Yemen and Libya.

• Least Developed Countries (LDCs): Sudan and Yemen
• Lower Middle Income Countries (LMICs): Egypt, Jordan, Syrian Arab Republic, Morocco, Tunisia, West Bank and Gaza Strip (OPT)
• Upper Middle Income Countries (UMICs): Algeria, Iraq, Lebanon, Libya, Turkey.

[1] The Foreign and Commonwealth Office and DFID worked collaboratively on developmental issues in MENA (DFID, 2014), and the priority areas for the new Foreign, Commonwealth and Development Office will be clarified once the HMG Strategy for MENA is finalised.

Summary findings

Evidence suggests that youth employment programmes are most successful when they form part of a comprehensive approach to reform and development, such as complementary interventions working on demand for youth employment, the supply of skills and policy reform (IEG, 2012; IMF, 2018; AfDB, 2012; Datta et al., 2018; WWINYE, 2020).

Although some meta-evaluations, most notably the 2017 evaluation of 107 youth employment programmes by Kluve et al., have found overall positive effects on employment and earnings from youth employment programmes, these effects were small. The Kluve et al. meta-evaluation covered skills training, entrepreneurship promotion, employment services and subsidies; they found effects to be greater in programmes targeting disadvantaged youth and in low- and middle-income countries. WWINYE (2020) confirms this finding. Kluve et al. also found that employment and earnings outcomes for women were marginally greater than those for men.

The evidence on the effectiveness of specific youth employment programme categories is mixed, complex and highly context-dependent. The table below summarises high-level messages related to evidence per intervention type and this summary should be read alongside the detailed findings in the body of the report. Across the board, our review found that the creation of jobs for young people at the expense of other groups, termed the displacement of non-targeted employment, to be a major problem limiting the net impact of youth employment programmes. However, in all cases, the sustainability of these effects is usually not monitored. Where it is, the effects tend to fall off quickly.

Strategic takeaways and core lessons

• The evidence base on what works remains limited and needs to be developed: While there is an increasing amount of literature on quantitative impact evaluations, evaluations rarely follow target populations over time, and therefore issues of sustainability of outcomes are not well addressed in the literature. Employment programmes often fail to report outcomes such as jobs created, or even how many young people were reached. For example, the 2012 thematic evaluation from the World Bank notes that 55% of Bank projects provide no information on youth as a beneficiary group (IEG, 2012). Finally, there is a limited body of evidence relating specifically to the MENA region (Kluve et al., 2017; ESAY, 2017).
• Addressing employment overall is critical: There is evidence that expansion of overall employment proportionately increases youth employment (Fox and Kaul, 2017) and helps to establish conditions for support within society for women’s economic empowerment (Gardiner, 2017). Moreover, economic growth and a stable political and macroeconomic context are important prerequisites for the successful promotion of youth employment (IEG, 2012). To target youth in low-growth, lower-income countries such as Egypt and Morocco, the literature suggests focusing on urban areas, employability skills and female empowerment.
• Targeting and engaging youth: There is a consensus across the literature that prioritising youth participation and engaging young people fully in the economic and social life of their communities yields stronger youth employment results (UN, 2018; UN, 2011; ADB, 2019). An ‘asset-based’ approach to working with young people means valuing their strengths, addressing their problems and ensuring that they have the capacity to transform their own circumstances (UN 2011; ADB 2019). Youth inclusion and empowerment is clearly important in MENA, where youth bear the burden of poor economic management (IMF, 2014). Inclusive governance approaches can also help promote the legitimacy of the government (OECD, 2016b). ODI (2018) notes that, for youth employment initiatives targeting refugees, programme designs should involve young refugees directly and build on what they are already doing to improve their circumstances.
• Shaping perceptions and job prestige: Effective communication with the public is a key success factor. MENA youth continue to value work in the bloated and less productive public sectors over private sector employment, so changing perceptions about job prestige is critical. A Jordanian job match scheme found evidence that ‘reservation prestige’ (that is, unwillingness to take on jobs perceived as less prestigious) is an important cause of unemployment among the country’s youth. 28% of participants rejected the offer to have an interview for low-prestige jobs, and even where job offers were made, 83% of applicants rejected the offer or left work soon after starting (Groh, 2014).

Programmatic and operational lessons

• Programme planning: Some positive findings are evident across the typology of youth employment programmes outlined in the Approach Paper for this review, but these are strongly shaped by For example, Kluve et al. (2017, p. 13) conclude that “…the ‘how’ seems to be more important than the ‘what’”, meaning that the details of implementation have a greater bearing on results than the intervention type. Studies highlight key common success factors, including coherence with broader donor efforts, the targeting of disadvantaged youth, providing incentives for youth participation, how youth are profiled and targeted, and creating incentives to motivate service provider performance. These findings are common across the MENA region. Kabbani (2019, p. 7) reinforces this point and concludes that “the inability to implement these policies and programs effectively and learn from mistakes has been a common failing across the [MENA] region”, exacerbated by complex approval processes for programme adaptation and course correction.
• Strengthen strategies and coherency of engagement: The existence of national youth strategies in the region is not correlated with youth employment outcomes and employment, and as such strategies appear not to improve the coherence of youth employment programmes and enterprise development initiatives (Kabbani, 2019). Lack of coordination across actors involved in economic reform and youth employment was a commonly cited problem, and the large number of ministries involved can also lead to duplication of effort and conflicting systems (see IEG, 2012; ESAY, 2017).
• Learn lessons from common factors of success: Datta et al. (2018) highlight success factors for youth employment programming, including programme designs that address multiple constraints on youth employment, understanding the context and the specific barriers faced by young people seeking employment, strong monitoring processes, incentivising private service providers, and involving employers actively in program design, implementation and evaluation. Targeting interventions (for example, based on demographic profile, industry or location) is another common recommendation, improving both results and cost effectiveness.
• Involving and incentivising the private and non-governmental organisation (NGO) sectors: Private sector and NGO implementation of youth employment programmes is widely found to improve the success rates of projects (for example, Fox and Kaul, 2017; IEG, 2012), particularly on refugee employment. However, this is not guaranteed. Kluve et al. (2017) found no systematic evidence in this regard in low-income settings. They found that including performance incentives to service providers correlated with better employment earnings outcomes (Kluve et al., 2017, p. 149, quoted in Hatayama, 2018).
• Choosing sectors: The sectoral composition of support to youth employment matters. For example, in rural areas, most youth are active in agriculture and self-employment, suggesting that a focus on agribusiness is important (IEG, 2012).
• Improve monitoring and evaluation of outcomes: Finally, even if sustainability and cost effectiveness of jobs created is not strong, practitioners should keep in mind broader outcomes that are less likely to be measured. Better monitoring and evaluation of programmes will help provide evidence in this regard.

Youth employment and instability

The literature suggests that any link between low youth employment and conflict or extremism is weak, indirect and context-based, rather than causal (Izzi, 2020; RUSI, 2015; Bricker and Foley, 2013; Cramer, 2011).

• Conflict: The evidence that youth employment programmes affect peacebuilding is scant, and a link between the two variables is often assumed, rather than evidenced (Izzi, 2020; Idris, 2016). Where there is a link, it was found to be complex and specific to context. Although youth unemployment can correlate with and may, at times, contribute to violence, it is only one of many relevant social and political factors, including corruption, injustice, humiliation and dysfunctional family relationships (Idris, 2016; RUSI, 2015). The research also showed that the link between instability and youth unemployment is due to the pressure that the cohort exerts on the total labour force, rather than the mere presence of a large number of young people (Bricker and Foley, 2013).
• Extremism: The evidence is clear that unemployment and poverty are neither necessary nor sufficient to push an individual into extremism or terrorism (RUSI, 2015). Although unemployment among more highly educated segments of the population correlated with higher probabilities of radicalisation (ibid.; Krueger and Maleckova, 2002; Bhatia and Ghanem, 2017), any link between unemployment and terrorism was found to be “indirect, complicated and … weak” (Krueger and Maleckova, 2002, p. 1; also Cramer, 2011; Hilker and Fraser, 2009).
• Migration: There is clear evidence that unemployment may result in migration, but the causal links are complex. The literature offers no evidence that any youth employment programmes reduce the likelihood of migration. Irregular migration tends to be triggered by loss of livelihoods, usually as a consequence of climatic stress or conflict (UNESCO, 2017; IZA, 2018). The decision to migrate irregularly is then shaped by the availability of regular migration routes, including the availability of visas, and weighed with factors such as differential income, cost of journey, a trusted network in the destination country, and the risk of detention or injury. In MENA, data shows that labour market formality is a firewall against irregular migration. Being employed in the public sector and being covered by the national security system both decrease the desire to migrate. Governance and democracy matter: youth are less likely to migrate when they report having a belief in the democratic ideal in their home country (IZA, 2018).

The MENA youth employment challenge

Almost half of the MENA region’s 448 million population is under 25 (World Bank, 2018b; UNICEF, 2019a) and five million youth enter the workforce each year (Drine, 2012). Due to a combination of macroeconomic factors with conflict, high rates of labour migration, poor governance, oil dependency, skills mismatches and a weak private sector, the MENA labour market has systematically failed to absorb young talent (Purfield et al., 2018; World Economic Forum (WEF), 2017).

Since 1991, the MENA region has had the highest youth unemployment rate in the world (ILO, 2020a), with around 23% of the region’s youth labour force (aged 15-24) unemployed (World Bank, 2020a; ILO, 2020a; ILO, 2020b; UNICEF, 2019a). Young people are five times more likely to be unemployed than adults (World Economic Forum, 2017). The nature and extent of the youth unemployment challenge varies widely across the region (World Bank, 2020c).  ILO data suggests that the current percentage of youth (15-25) who are unemployed ranges from 18% in Lebanon, to 50% in Libya[2].  Across our specific cohort of 13 ODA-recipient countries (see Section 1), the average is higher than the MENA wide figure,  with nearly 30% youth unemployed, contrasted to a total unemployment rate in those countries of just 9%.^[3] Unemployment disproportionately affects those more highly qualified in the region: 30% of unemployed youth are graduates, which is a “marked contrast to global patterns” (WEF, 2017, p. 4).

This analysis  is based on the most recent formally reported data and does not take into account the impact that the COVID-19 pandemic and other more recent shocks have had on unemployment.^[4] These shocks are likely to have exacerbated the dire picture, particularly in Lebanon.

Unemployment is particularly acute among young women (see also UNICEF MENA ADAP, 2019). On average, it takes young men up to three years to make the education-to-work transition, and many young women are encouraged not to enter the workforce at all (ibid.). According to the International Labour Organization (ILO), over 50% of women in MENA are NEET, compared with under 20% of men.

Economic trends

Low economic growth

There are significant regional variations in GDP per capita across MENA. Gulf states are the wealthiest in the region: for example, the UAE’s GDP per capita average is over 170% of the global average. The Levant follows but with a wide gap. Lebanon tops that region with a GDP per capita value of just 40% of the global average. North Africa in general lags further behind (OECD, 2016a). The data does not take into account current events, such as the impact of COVID-19 and the currency devaluation in Lebanon.

Data on GDP also shows low growth across the cohort (World Bank national accounts and OECD National Accounts data files, 2020). While the financial crisis had a limited impact, the data clearly outlines the impact of conflict on affected countries, particularly Yemen, Libya, the OPT and Iraq.^[5] The data also shows significant downward trends in GDP in Morocco, Tunisia and Libya following the Arab Spring and conflicts in Libya and Iraq which inhibited regional trade. Finally, the data suggests that GDP is most volatile in LMICs.

Commentators suggest that MENA was somewhat shielded from the impact of the 2008 global financial crisis, in part because of its low integration into international capital markets and high public spending (Neaime, 2019). These same factors, however, mean that growth in MENA has been slow and unequal since 2009 (Purfield et al., 2018; OECD, 2016a).

A weak social contract

The historic reliance on oil, oil services and natural mineral industries in key countries of interest, such as Egypt, Iraq and Morocco, and the use of large public sectors to generate employment, has formed the modern social contract of various MENA countries: state legitimacy has rested on government ability to deliver services, subsidies and jobs (Assad et al., 2018; Chen and Harvey, 2017; Dadush, 2018).

Falls in oil prices since the 1980s have placed added pressure on government budgets and limited economic investment in the countries reliant on this revenue, such as Iraq and Egypt (Chen and Harvey, 2017; Kabbani, 2019). Net energy importers such as Egypt, Tunisia, Lebanon, Jordan and Morocco have been unable to boost their manufacturing and service industries sufficiently to create enough jobs for their growing populations, resulting in large diasporas across the region and in Europe (Dadush, 2018; Raz, 2019; WEF, 2017).

MENA countries are therefore trapped in a cycle of high reliance on government, low public investment, low tax revenues and low oil prices  – factors which limit governments’ fiscal headroom. High spending on fuel subsidies and public service jobs reduces government ability to enact fiscal policies that support growth (including on social protection, health and education: IMF, 2018). Barriers to reform in the private and public sector have further contributed to this crisis, while red tape has hampered investment (Kabbani, 2019).

While the countries of the region on average place 107 (of 190) on the World Bank’s 2020 Ease of Doing Business ranking, the ICAI cohort performs much worse. There are wide variations within the cohort, with Turkey, Morocco, Jordan and Tunisia outperforming conflict-affected Yemen, Libya, Syria, Iraq and Sudan.

Entrenched reliance on public sector jobs

Economies in the MENA region are characterised by underdeveloped private sectors. Total firm revenue accounts for only 40% of total regional GDP, significantly below the OECD average of 59% (OECD, 2016a). Bureaucratic barriers to doing business and accessing finance, and widespread corruption and cronyism, have inhibited private sector growth (Haidar, 2016; Kabbani, 2019; OECD, 2016a; UNICEF MENA ADAP, 2019). When combined with constraints on the ability of governments across the region to create more public sector jobs, this has resulted in chronic high unemployment in recent generations.

Perverse incentives also play a role in entrenching the reliance on public sector jobs across the region. High wages and benefits have made public sector jobs more lucrative than private sector employment (Assad et al., 2018; Chen and Harvey, 2017). Cultural and social norms outlined in Section 2.3 (Structural and intergenerational dynamics) also place more value on public sector jobs, incentivising young people to ‘wait’ for lucrative jobs or take up informal employment while ‘queuing’ for a public sector post (Dadush, 2018; Deloitte & Touche, 2018; Kabbani, 2019; Silver, 2017). This trend is also mirrored in the formal education system, where skills needed for private sector jobs are less valued, leaving young graduates under-skilled and further exacerbating the lack of youth employment (Dadush, 2018; Deloitte & Touche, 2018; Haidar, 2016; WEF, 2017).

Despite relatively high levels of education (Haidar, 2016; UNICEF, 2019b), weaknesses in the business environment and the financial sector (in terms of both lending ability and financial infrastructure) have meant that entrepreneurship is a difficult avenue for young people to pursue. Most private business in the region are small and informal: only 3.4% of MENA entrepreneurs say that their business employs others (UNICEF MENA ADAP, 2019). ‘Own account workers’ are more likely to be in precarious and vulnerable working environments (ibid.).

Political and economic instability has led to uncertainty and weakened markets

UNICEF (2018) describes 60% of the region as fragile. Violence in Syria, Iraq and Yemen, resulting in internal displacement and regional migration into countries such as Lebanon and Jordan (Dadush, 2018; Kabbani, 2019), has reduced investment and weakened tourism in home and host countries (OECD, 2016a).

Political and economic instability is not only linked to conflict. The oil-rich Gulf is consistently dependent on foreign labour and oil revenue to balance budgets, while also suffering from high rates of local unemployment (due to skills limitations or a desire by citizens to avoid work in migrant worker-dominated sectors: Dadush, 2018). With high reliance on migrant workers, Saudi Arabia is the world’s second-largest source of remittances and attracts labour from Lebanon, Jordan and other countries in the region. As a result of this, wages earned in Saudi Arabia are then remitted across the region (Dadush, 2018; Kabbani, 2019; WEF, 2017).

Social trends

Structural and intergenerational dynamics

The underlying causes of youth economic exclusion in MENA are institutional. They include the systems that shape outcomes, incentives and behaviours, whether of youth, their parents, employers or others, in the spheres of education, employment, marriage and housing. The legacy of high wages and security for government employees means that many young people continue to queue for jobs in this sector (WEF, 2012). Younger generations’ attitudes to the public sector largely match the existing social contract, with 82% of young people in the Gulf countries, 80% in North Africa and 71% in the Levant agreeing that it is a state’s responsibility to provide job opportunities (ASDA’A BCW, 2019).

Coupled with the economic trends outlined above, the importance of connections, termed ‘wasta’, inefficiencies and poor targeting in public spending and lack of voice are among the barriers to inclusion. Between 60 and 90% of respondents to a Gallup survey conducted in 2013 in countries across the region think that knowing people in high positions is critical to getting a job (IMF, 2018).

Facing poor economic prospects and inadequate income, youth in MENA are unable to marry, afford to live independently or support a family. This stalled transition from youth to adulthood is also referred to as ‘waithood’, a time of relative inactivity and uncertainty that can fuel frustration and helplessness among youth (Brown, 2014; WEF, 2012). In MENA, men aged between 25 and 29 have the lowest marriage rate within the developing world (WEF, 2012). Based on extensive research in Egypt and the Middle East, Singerman (2007) identified four factors fuelling marriage delay: a demographic youth bulge, high youth unemployment rates, high marriage costs and low wages.

Young people therefore remain dependent on their families or on relatives working abroad, unable to move on to the next phase of their lives. Chamlou (2016) argues that contradictory policymaking underpins the ‘MENA Paradox’: a system of heavy state subsidies unintentionally encourages and sustains one-income families.

In Jordan, as with other parts of MENA, parents’ expectations are a key factor in limiting employment choice for young women, as only certain types and locations of work are considered socially acceptable. The public sector is generally more acceptable than smaller, less structured private firms, which parents perceive as offering less security and protection. The lack of available jobs in the public sector tends therefore to disproportionately affect women (Brown, 2014). Most young Jordanian women in Brown’s study expressed an intention to work, which suggests that low employment is less a result of internal cultural or social views discouraging economic participation, but rather a result of external barriers, including by family or employers, or other constraints such as security and access to child care.

Gender and unemployment

The marginalisation of young people is especially felt by young women, who also face gender discrimination in the workplace. The Middle East has been consistently underperforming in the Global Gender Gap Index, with a score of 61% in 2020 compared with the 69% global average (World Economic Forum, 2019). While significant progress has been made in closing the gender gap in educational attainment, structural discrimination against women in the workplace – and a preference for male employees – has left female employment low across the region (Assad et al., 2018; Haidar, 2016; UNICEF, 2019b; World Economic Forum, 2019).

Women’s workforce participation rates remain low in MENA, and those women who do work outside the home face a wide range of legal, institutional and cultural challenges to their participation in the labour force (IMF, 2018; Chamlou, 2016; Kabbani, 2019). This contrasts with global trends, where high female educational attainment and marked reductions in fertility have seen more women enter the workforce (ibid.). In MENA, over 80% of working women are in employment which is vulnerable. Women also spend up to three hours more each day on household labour than men and are less likely than men to have a bank account (58% compared with 65%: S4YE, 2015).

According to Chamlou (2016), MENA governments have promoted progress for women in education and health, while also safeguarding perceived traditional cultural norms and male dominance through discriminatory laws and policies that limit female economic participation. Marriage – and particularly a spouse’s attitude toward women’s work outside the home – affects women’s economic activity more significantly than having children. The World Bank’s Women, Business and the Law report (2020) notes that while some progress has been made (from a low baseline), women in MENA still face significantly more legal obstacles relative to men in the region and relative to women in other parts of the world. The current overall MENA index score of 49.6 suggests that women have just under half the rights enjoyed by men in the areas covered by the index.

NEET rates for young women are particularly stark. In the OPT, 25% of young men and 38% of young women are considered NEET, and in Egypt, the corresponding shares are 17% and 41% (IMF, 2018). For women, the NEET gap persists over time. NEET rates of young women in their late 20s and early 30s correspond very closely to their status in their late teens and early 20s (ILO, 2019).

Social vulnerability

Despite the relative wealth of some countries of the region, 26% of MENA populations still live in poverty as measured in multi-dimensional terms, with higher rates in rural and disadvantaged areas. Social safety nets (SSNs) are limited in the region by international comparison. Except for the OPT (which spend around 5% of GDP on SSNs), most MENA countries spend just 1% of GDP. Targeting of SSN programmes has also been poor: of the 11 MENA countries for which data is available, on average only 21% of all spending on social protection (social assistance, social insurance, and labour market programmes) goes to the poorest 40% of the population and 41% is captured by the richest 20% (IMF, 2018). Coverage of the population in the lowest income quintile can be as low as 12 to 20%, with often limited awareness of such programmes among targeted groups (IMF, 2018).

The Social Exclusion Index data from Algeria, Egypt, Lebanon and Tunisia indicates that the share of young people experiencing social exclusion is highest in Tunisia (46.7%), followed by Algeria (43.4%), Egypt (42.1%) and Lebanon (33.2%) (Backeberg and Tholen, 2017).

Disability

Youth with disability are among the most marginalised and more likely to face severe social, economic and civic disparities compared with their peers. Young women with disabilities face both disability and gender-based discrimination. Article 27 of the Convention on the Rights of People with Disability states that all people with a disability, including youth, have the right to work (World Bank, 2010; WRC, 2017). All countries in our cohort have signed the convention, and many MENA countries have specific legal provisions pertaining to employment for people with disabilities. Egypt, Lebanon and Tunisia also have employment quotas for people with disabilities (ranging from 1 to 5%) (World Bank, 2010). That said, youth with a disability are more likely to be in the informal labour market and not easily reached by reforms in regulatory frameworks or affirmative action.

Youth perceptions and priorities

Young people in MENA rate economic hardships among their most significant concerns in the region. In a 2019 region-wide survey into the “hopes, fears and aspirations of Arab youth”,^[6] young people rated the rising cost of living (56%) and unemployment (45%) as their top two concerns (ASDA’A BCW, 2019, p. 10).^[7] This was a marked change from previous years when concerns focused on civil unrest, lack of democracy, Da’esh and terrorism.

Young people across the region are mostly satisfied with their level of skill and the quality of their education, with some variations between countries. Iraq and Morocco, for example, report significantly lower levels of satisfaction, with 24% and 17% respectively (Raz, 2019). This satisfaction is not matched by employers, who say that graduates’ interest in generalised areas of study limits the uptake of graduate labour (Haidar, 2016). While many young people reported being unprepared for the job market by their place of study (ASDA’A BCW, 2019), they feel that their skills and determination make up for this lack of direction.

In a survey on attributes ranked when picking a job, young people selected ‘passion’ as the most valued attribute at 44%, followed by competitive salaries (9%) and working for a well-known organisation (7%) (Bayt.com, 2017). A recent Programme for International Student Assessment survey also showed that teenagers in MENA show strong preferences for traditional career routes in sectors such as medicine, teaching and engineering, rather than careers in emerging or technology-based areas such as artificial intelligence and the digitisation of the workplace. This further exacerbates the issue of a lack of technical expertise and career direction in the schooling system (Abd El-Galil, 2019).

Socially conservative attitudes on female labour force uptake remain, with young men in the MENA region breaking a global trend by being more conservative than their older counterparts (El Feki et al., 2017). This can be partially explained by the lack of general employment opportunities in the region, which makes them more hostile to perceived competition while also being expected to shoulder the responsibility of being breadwinners for their households. Three out of four men in the region state that a woman’s most important role is in the home, and while older women in the region largely mirror this view, younger women consistently show a wish for change and greater equality of opportunity (El Feki et al., 2017).

Young people in MENA also feel politically disenfranchised (UNICEF, 2019a; Raz, 2019). Civic disengagement in MENA has been cited in multiple reports, including the Arab Human Development Report (2016) and Mercy Corps (2012). The 2019 survey outlined above sampled 2,400 youth across MENA, and large numbers reported a lack of interest in politics (Raz, 2019). Interest in politics was particularly low in Algeria (15%), Tunisia (17%), Libya and Iraq (both 21%), in contrast with the highest-ranking countries: Lebanon (31%), Yemen and the OPT (both 29%)  (ibid.).

The systemic lack of work opportunities has a widespread effect on young people’s migration patterns in the region. According to 2019 data from the annual ASDA’A BCW survey, over two-thirds of young French speakers in North Africa say they want to migrate to Europe, while 60% of youth in Egypt, 48% in Yemen and 45% in Sudan say they want to emigrate to a Gulf country (Raz, 2019).

[2]  Source: ILO 2020c and 2020d; accessed on 30 June 2020

[3]  Youth unemployment data in MENA can be unreliable due to variations in the way that national estimates are gathered by different countries (Kabbani, 2019 and العربي للنقابات الاتحاد 2017). In the Gulf region, for example, states only account for Gulf nationals in their survey data despite hosting sizeable diaspora communities. Furthermore, the data reported could be years old and may not reflect current realities in MENA, particularly in conflict-affected countries or countries under political or economic stress. For example, current youth unemployment rates in Syria and Lebanon are likely to be more acute.

[4] See, for example, recent figures that estimate a youth unemployment rate of over 23% in Lebanon; Labour Force and Household Living Conditions Survey 2018-2019, Lebanese Republic, 2021, link.

[5] Data reported by the Syrian government also showed a 66% decline in GDP between 2011 and 2018; Number of population existed in Syria according to estimates of their number in mid years 2011-2018, Central Bureau of Statistics, Syria,  link.

[6] This covers all of our cohort except for Turkey.

[7]  In 2012 and 2013 the “rising cost of living” was the primary concern, followed by government corruption and the economy.

Youth employment: what works?

Interventions to increase youth employment straddle a broad spectrum of programme types to address either the quantity and quality of jobs available (the ‘demand’ side) or the quality of the workforce (the ‘supply’ side). There are large numbers of these programmes, both worldwide and in MENA specifically, and they vary in scope: Egypt alone, for example, implemented over 180 youth employment projects between 2007 and 2014 (Kabbani, 2019).

Analytical framework and availability of literature

In Figure 1 we provide an outline of a broad typology of youth employment programmes developed in the Approach Paper, which will be used across this review. To ensure a focus on the most relevant elements of the typology to youth employment programming, the review will focus on active labour market programmes and direct employment or employability skills. It will not encompass broader macroeconomic support or education programmes. These six areas form the backbone structure of this literature review. Within each section, we outline the balance of evidence on what works, the factors (contextual, structural and others) that enable or impede programme success, evidence of sustainability and specific lessons relevant to the design and implementation of projects.

Limitations

The amount of quality evaluative material available on each element of the typology varies. While quantitative impact evaluations are increasing in number, evaluations rarely follow target populations over time, and so sustainability of outcomes is insufficiently documented. Moreover, employment programmes often do not report outcome indicators such as the net number of jobs created, or even the number of youth reached. For example, the 2012 thematic evaluation from the World Bank notes that 55% of Bank projects provide no information on youth as a beneficiary group (IEG, 2012). Finally, there is not a large volume of evidence pertaining specifically to the MENA region (Kluve et al., 2017; ESAY, 2017).

Even where evaluations are available, drawing general conclusions from the findings is not straightforward (see, for example, Ismail, 2018a). There is nonetheless a large body of literature on the subject, including several meta-evaluations or systematic reviews that include significant amounts of evaluative material. Examples include IEG (2012), which encompasses 38 impact evaluations, the most comprehensive and regularly cited Kluve et al. (2017), which covers 107 evaluations, and the 2017 Fox and Kaul study which covers an unspecified number of evaluations. We note that many studies refer to the Kluve et al. findings, often using the same language, so the number of references to any specific finding is not necessarily a good indicator of independent agreement.

Creating an enabling environment

This section considers evidence on programmes to promote youth employment through creating an enabling environment. This group of activities includes regulatory reform, reform to wage policies, access to work permits, access to credit and use of special economic zones. This section also considers the specific needs of women and refugees regarding programming, policy and regulatory reform.

Given the breadth of topics under discussion, this section can only summarise the core lessons within each programme area. Overall, the evidence base on how enabling environment programmes can promote youth employment is limited due to the gap in data collection on the topic. Several reports from the World Bank (IEG, 2012) and IMF (2014) note the need to systematically study the impact of reform programmes on youth employment (see Sustainability, below).

Improving rule of law and business regulation implementation: Youth employment is a subset of the overall employment and earnings challenge in low-income countries, namely an oversupply of workers relative to the number of firms (Fox and Kaul, 2017). Job creation is predicated on the growth of small and medium-sized enterprises, but impediments to private sector growth are mainly experienced by small firms in the MENA region. These stem from the often discretionary implementation of business regulation (in favour of large politically connected firm owners) rather than the absence of regulation, meaning that corruption and anti-competitive practices coupled with opaque laws act together to restrict private sector growth (UNDP, 2013).^[8]

The literature reiterates the need to reduce bureaucratic discretion and the reliance on political or personal connections, or ‘wasta’, in Arabic, as both factors disadvantage those who lack government connections, chiefly small firms or new entrants to the market (IMF, 2018). With large modern firms being responsible for the majority of new jobs (ibid.), successful reforms include support to firm entry and expansion by removing legal and administrative barriers to, and incentivising, foreign direct investment and improving access to the judiciary, land or industrial zones (World Bank, 2015). However, policies that privilege politically connected firms still persist despite their broader economic cost (Chaaban, 2013; World Bank, 2015).

Minimum wages and strong labour market regulation incentivise informality and harm youth employment: High minimum wages negatively affect employment and minimum wages in MENA tend to be higher than in other developing countries, such as those in Eastern Europe and South America (Chaaban, 2013). Coupled with relatively high labour taxes^[9] and employment taxes, this incentivises small firms to remain in the informal sector (World Bank, 2013; Chaaban, 2013). With public sector wages also exceeding private sector wages and disconnected from productivity, these wage policies have a damaging effect and young people end up ‘queuing’ for public sector jobs (IMF, 2018).

Strict labour market regulations and contracts disadvantage youth while protecting ‘insider’ workers (World Bank IEG, 2012; World Bank, 2013). In South America, the World Bank (2013) found that the unemployment effect of stricter regulation is almost twice as large for youth than adults. Put simply, strict minimum wage laws and regulatory protection for workers restrict youth employment and push youth into precarious or informal employment (ibid.).

Legislative minimum wage exemptions for youth increase employment (World Bank, 2013). Although this has been highlighted in Bank reports on several occasions, the Bank’s Independent Evaluation Group (IEG) found that “none of the analytical work discusses how alternative regulations should be formulated or what their fiscal impact would be” (emphasis added; IEG, 2012, p. 38). As such, there is no clear evidence on how to approach deregulation to promote youth employment, from either a political or an economic perspective. More evidence is critical because the impact of liberalising labour contracts can be mixed: it may also fragment the labour market, increase turnover of workers and limit possibilities to successfully transition to permanent positions (World Bank, 2013).

Access to credit may support youth employment: Obstacles to accessing credit stifle investment and business growth. MENA countries systematically underperform on access to finance (World Bank, 2015). Young people are, in particular, underserved by financial institutions, including banks, credit unions and microfinance institutions (UNDP Arab HRD, 2017). Collateral requirements disproportionately affect women, as assets tend to be in their spouses’ names (ILO, 2018a).

Evaluative material suggests that cash grants to finance start-up costs provide short-term boosts to employment, but their effects do not seem to last long (Card, Kluve and Weber, 2015). In Yemen, a youth-lending facility was evaluated and showed positive results. The facility emerged from a partnership between Silateck and Al-Amal Microfinance Bank and repayment rates were higher among youth than older clients. Similarly, Kiva, a microlending platform, mobilised USD 5.5 million in funding for 3,900 Arab businesses. It enjoyed a 98% repayment rate and helped create 30,000 jobs (UNDP Arab HRD, 2017).

The literature suggests, however, that the type of loan and size of firm affect the success of access-to-credit programmes. A review of the impact of microcredit on business growth suggests that, below a certain level of income, borrowing may lead to a poverty trap as borrowers never reach their efficient size to be able to repay the loan and generate a surplus. This has the potential to create a debt trap, because borrowing can continue in perpetuity (Liu and Roth, 2017). (Business) size matters: “firms at very low or very high business sizes grow faster in the presence of a lender than in his absence…whereas firms at the intermediate business sizes may not grow at all in the presence of a lender” (ibid., p.24; also World Bank, 2011; see also SAYE, 2015). Finally, evidence from the UK suggests that to avoid a debt trap, access to credit must go hand in hand with financial education (Action for Children, undated). Section 4      examines the issue of credit for entrepreneurship in more depth.

Limited evidence that special economic zones (SEZs) support youth employment: SEZs are zones created to attract investment and promote industrial activity through financial and regulatory incentives. In MENA, large firms in politically connected industries are more likely to be based in an SEZ. Through their connections such firms are also able to shape tax policy to their favour, entrenching their preferential treatment and reducing competition (World Bank, 2015). The literature also questions the broader impact of SEZs on economic transformation. Even where zones have successfully generated investment, evidence suggests that their overall impact on growth is temporary (UNCTAD, 2019) and that the broader economic benefits of SEZs are harder to gauge as they operate “as enclaves, with few links to local suppliers and few spillovers” (ibid., p. 129).

Although SEZs can be viable sources of jobs, the ILO finds that “they do not present a solution to unemployment” due to, among other factors, high turnover rates and low investment in skills. The ILO also notes the difficulty in measuring whether investments in these zones are, indeed, additional or would have occurred even without the existence of the SEZ (ILO, 2017a; IMF, 2017).

Work permits and mobility support employment formality: There is significant labour mobility in the region: the Gulf states, Lebanon and Jordan all experience heavy inward migration of low-skilled labour coupled with significant overseas migration of their educated population (9% in Jordan, 14% in Lebanon) (UNDP, 2016). This issue is particularly relevant to the large refugee populations in the Middle East. Work permit restrictions, limited recognition of qualifications and a lack of harmonisation across visa systems affects youth labour force participation (ibid.; see also Jadaliyya, 2017).

Specifically, the United Nations Development Programme (UNDP) (2016) cites the lack of transparency and harmonisation in visa systems and in the duration of visa validity as an additional barrier to labour mobility and notes that an Arab citizen needs on average around 16 visas to travel to all countries in the region. It calls for a business travel card system as implemented by the Asia-Pacific Economic Cooperation region initiated in 1999 to allow multiple short-term visits to member countries over three years as well as a single stay of up to three months (ibid.). With already high rates of unemployment in MENA countries, however, such moves may be politically unpalatable.

Syrian and Palestinian refugees in Lebanon and Jordan are particularly affected by work permit restrictions, though this problem is understood to be due to politics rather than any technical reason (see, for example, ILO, 2018). In Lebanon, for example, the ILO said that what is lacking is jobs “with decent working conditions in sectors that Palestinian refugees are legally allowed to work in” (emphasis added; Kherfi et al., p. 18), and not the quantity of jobs. This situation drives the refugees to dependency on UNRWA (the UN agency established to provide support to Palestinian refugees), to undertake precarious self-employment or into exploitative working conditions (ILO, 2018b). Similarly, one 2017 ILO study found more Egyptian than Syrian nationals employed in the Jordanian workforce (ILO, 2017d). This reinforced the view that the primary impediment to refugees accessing jobs was political, rather than a shortage of available positions.

Creating an enabling environment for women through improving social protection and reducing legal differences based on gender: Legal, institutional, religious and discriminatory barriers greatly impede female access to employment: women face travel obstacles, difficulty obtaining identification and complications registering a business, as these activities sometimes require spousal permission (UNDP, 2013). According to IMG (2018), 18 of the 30 countries with ten or more legal differences based on gender are in the MENA region.

The literature suggests that more young women would be willing to work and remain in the labour market if social protection were extended to cover flexible and part-time work (UNDP, 2016; UNDP, 2013). Women’s employment would also be advanced by raising the minimum marriage age, mandating maternity benefits, increasing women’s access to property, reforming guardianship laws and repealing pension laws that mandate earlier retirement for women than men (Chaaban, 2013). Despite laws protecting equal pay for women, enforcement of anti-discrimination laws is uneven. Inequalities in social (that is, non-wage) benefits, such as child and family allowances, also undermine pay equality as such allowances are often paid directly to men (ibid.).

Creating an enabling environment for refugees through flexible financing and work permits: According to the International Monetary Fund (IMF), countries emerging from conflict or hosting substantial refugee populations require specific and tailored support to promote employment. In an International Finance Corporation (IFC) review, the organisation outlined three “critical enablers” for refugee employment: 1) flexible financing; 2) cross-sector partnerships; and 3) investment information. The evaluation noted that only 1% of grant-based philanthropic investment in the Sustainable Development Goals between 2016 and 2018 was designated for migrants and refugees. This funding also tends to prefer larger and more mature organisations, even when the “challenges refugees face can at times be better addressed by smaller, more nimble entrepreneurial organisations” (IFC, 2019, p. 12).

A lack of work permits for young refugees pushes them into the precarious and low-paid informal economy. In Jordan, for example, in 2014 only 6,000 of 160,000 Syrian refugees estimated to be working in the country had a work permit (Jadaliyya, 2017). To overcome this, the Jordan Compact focused on easing barriers to obtaining work permits (including high fees), and helping refugees formalise existing and new businesses (ibid.).

Refugee women face an even greater burden: in addition to the barriers described above, they must also overcome social, security and cultural pressures, specifically within camps (ILO, 2018a). Across the board, the IFC recognised that programming to improve refugee employment depends upon political leadership and a supportive policy environment – in particular, a regulatory framework that enables refugees to access employment (IFC, 2019).

Youth strategies have limited impact: In contrast to Europe and Central Asia, where high youth unemployment rates have been reduced, national youth strategies in MENA do not correlate with improved youth economic outcomes (Kabbani, 2019). This is predominantly understood to be due to the disconnect between on-paper strategies – often designed in ministries of youth and sports which lack resources and material political clout – and the political appetite or technical ability to operationalise necessary legal reforms, employment policies and enterprise development strategies (ibid.).

Sustainability – insufficient evidence

There is broad recognition across the literature that high rates of youth unemployment are linked to underlying economic structures. However, programmes that attempt to address structural impediments have not been systematically evaluated for their effect on youth employment (World Bank IEG, 2012). This means that we found limited evaluative evidence on which enabling environment reforms most sustainably promote youth employment, and how such reforms could be designed.

Even after the World Bank IEG study, which reinforced the need to monitor the effects of reform efforts on youth employment, subsequent Bank reports (for example, in 2015, on “unleashing the employment potential of the Middle East and North Africa”) mention ‘youth’ a mere four times, and women only 13 times over several hundred pages. The literature suggests that multilateral organisations understand the importance of broadening the evidence base. According to the IMF (2014, p. 84), with youth bearing a disproportionate share of the costs of poor economic management, “youth interests can no longer be treated as an ‘add-on’”.

Lessons learned – promote integrated programmes and monitor their effects on youth employment

Given the relatively limited evidence base available on the impact of specific programmes to promote an enabling regulatory environment on youth employment, it is difficult to identify clear lessons. As described above, there is broad agreement that reducing minimum wages and employee protection would benefit youth, as would supporting access to credit and labour mobility, improving business regulation and rights, reducing preferential treatment for politically connected firms, and promoting better governance. All this needs to be underpinned by an integrated effort – by national governments and donors – to support growth and systematic monitoring of the impact of broader reform efforts on youth employment.

Underpinning these reforms, the literature recognises that creating an equitable enabling environment for youth employment relies upon both 1) improved public communications and 2) addressing biases and norms regarding ‘good’ jobs and the economic role of women. The IMF, for example, highlighted the need for a “comprehensive communications strategy” and greater information on the costs of subsidies and benefits of reform to ensure public buy-in for these changes (IMF, 2014).

Communication campaigns are important to increase women’s participation in the labour force, as men’s views of women’s rights matter. Gallup reports found that the more men support women’s workforce participation, the more women are likely to work in professional jobs (WEF, 2012). The report also found that men’s views of religious laws have “no correlation whatsoever to how they regard women’s equality”. Rather, “the more men are happy, employed and educated, the more they support women’s rights” (ibid., p. 13). Finally, a review of a television programme broadcast between 2013 and 2014 to promote entrepreneurship among young people shows the potential role the media can play to address socio-cultural barriers and attitudes that constrain women in the workplace (ILO, 2018a). By embedding educational content into an entertainment format, the programme had a significant impact on viewers’ views on gender roles, and specifically reduced discriminatory views held by men against women entrepreneurs (ILO, 2018a).

Communications campaigns are important for addressing the mismatch between aspirations, expectations and what the labour market offers: Young people’s aspirations and expectations, beliefs and attitudes about work may shape their employment outcomes even before they start working. Employers will also have expectations of readiness and experience, and assumptions that may cause them to avoid hiring youth, who are seen as riskier employees (S4YE, 2015). The extent to which this mismatch explains unemployment or prolonged job search, especially in low- and middle-income countries, is not yet well known (Fox and Kaul, 2017). In Jordan, USAID developed a toolkit to explore existing stereotypes and help change parental attitudes. Initial assessments of the programme found activities to be effective in bridging the gap between parents and their offspring, particularly when it comes to enhancing communication and supporting young people’s planning for jobs and the future (USAID, 2012).

Employment subsidies

Using subsidies to promote the employment of targeted groups is contentious. It is a political decision and one which donors do not often support. The UK aid portfolio does, however, include some specific, and high-cost, wage subsidies in certain contexts. Subsides are expensive, unsustainable and do not result in net employment gains. Where individuals benefit, this comes at the cost of other employees being displaced. They are also not sustainable as, once subsidies end, the employment effects have been found to disappear. The process of removing subsidies is difficult and can cause political unrest (AfDB, 2012; S4YE, 2015; WWINYE, 2020d; Broecke, 2013).

Despite this evaluative evidence, employment subsidies are used frequently by governments to manage the unemployment of specific groups. The African Development Bank (AfDB, 2012) notes that, in response to the Arab Spring, the Egyptian government hired an additional one million civil servants, increasing the wage bill by 13% in 2011. Libya, Morocco and Jordan also increased salaries for civil servants and expanded subsidies as part of their response to the Arab Spring (ibid.).

Subsidising the employment wage bill supports target employment, but does not increase net total employment: Across the board, evidence suggests that subsidising employment, either through direct subsidies or by reducing the employer’s social security contributions, has a positive impact on employment in the target group, but not necessarily on net total employment (S4YE, 2015; WWINYE, 2020d; Broecke, 2013). This impact appears higher for middle-income than high-income countries (WWINYE, 2020d; Groh, 2012). A regional example is Turkey’s 2008 Employment Package, which combined a tax subsidy for employing youth and cuts to social security payments. This resulted in a 23% increase in formal employment for young men. A Yemeni example of subsidising young interns showed increased employment among beneficiaries two months later (Fox and Kaul, 2017). Another impact evaluation of voucher-based subsidies to target rural youth in South Africa found a 15 to 25% increase in employment among target groups (Hatayama, 2018). The use of vouchers, however, is not always an effective route. In Turkey, for example, although the wage subsidies had positive effects, voucher-based programmes did not (Grimm and Paffhausen, 2015, p. 12).

Size of effect is contentious: Some consider the evidence that wage subsidies increase employment to be inconclusive (Ismail, 2018b) or as having a small effect (Kluve et al., 2017). Many authors point out that the increase in employment is not necessarily a net increase: it leads to the displacement and substitution of workers, as individuals in the target population are employed at the expense of others (Broecke, 2013; O’Higgins, 2017). Displacement can also take place between formal and informal roles. In Turkey, much of the employment generated was simply a formalising of previously informal jobs: in other words there was no net increase in job creation (Fox and Kaul, 2017). A further criticism of wage subsidies is that they subsidise jobs which would have been created anyway, rather than leading to new employment (referred to in the literature as ‘deadweight cost’) (Broecke, 2013; Datta et al., 2018; O’Higgins, 2017). Youth wage subsidies can also have a perverse effect by increasing the incentive to leave education or increasing the stigma associated with receiving benefits (O’Higgins, 2017).

Other subsidies may promote formal employment: In Ethiopia, a transport subsidy did not impact employment overall but, contrary to wage subsidy evidence, it did increase formal employment (Fox and Kaul, 2017). A similar subsidy in Bangladesh resulted in increased employment and earnings. Both effects were understood to be achieved through the displacement of other workers (ibid.).

Sustainability – mixed evidence

Regardless of approach, there is mixed evidence as to whether these effects are sustained. Kluve et al. (2017) reviewed 17 wage subsidy programmes in high- and middle-income countries and found that the probability of employment increased beyond the subsidy duration (p. 165). Others agree with this finding (WWINYE, 2020d). The South African example found the effects to be sustainable, lasting at least two years after the scheme ended (Hatayama, 2018), although the number of beneficiaries was small (Ismail, 2018b).

However, many others disagree (Fox and Kaul, 2017; Datta et al., 2018; O’Higgins, 2017; Groh, 2012). A subsidy in Sri Lanka did not have a sustained effect on employment. A study in Chile found that wage subsidies increased employment of vulnerable youth by 13% within the initial six months, but this figure dropped to just 3% after 18 months (Ismail, 2018b). In the MENA region, a Jordanian voucher programme to encourage female youth employment increased employment by 40%, but this work was informal and was not sustained: the effect was not significant four months after the voucher period (Groh, 2012).

The Tunisian national subsidy programme for employing university graduates (Stage d’Initiation à la Vie Professionelle, SIVP) supported 15,000 graduates in 2004 and 45,000 by 2011 (Broecke, 2013). The programme increased the probability of employment in the private sector (reducing joblessness among university graduates by 8%) but also reduced the chances of permanent contracts among young programme participants. The lack of candidate profiling was found to have resulted in the programme subsidising jobs that would have been created either way (causing large ‘deadweight’ effects). However, ESAY 2017 found that the programme did have some sustained effects – beneficiaries were less likely to be unemployed and significantly more likely to obtain a private sector job upon completion.

Many examples do not benefit from longitudinal monitoring, meaning that the question of sustainability is not addressed. WWINYE (2020d) notes that there is no evidence of improved duration or ability to retain jobs after the programmes. On balance, the available evidence suggests that where subsidies promote employment, the effects are reversed when the job subsidy ends.

Lessons learned – combine wage subsidies with skills training to improve employability

There is broad agreement that combining wage subsidies with skills training can help improve employability among youth (Kluve et al., 2017, p. 165; Hatayama, 2018; WWINYE, 2020d; O’Higgins, 2017). However, regional evidence does not concur with this. A Jordanian voucher scheme, linked with soft skills training, underwent a randomised trial and the results showed that soft skills training had no impact on employment (Groh, 2012).

The ILO’s review of wage subsidies suggests that programmes with a medium duration (between six months and two years) are most effective. Shorter programmes do provide opportunities for employers to gain information about young workers but are not long enough to improve skills in young people and increase their chances of employment (O’Higgins, 2017).

This mixed evidence highlights the importance of good design in wage subsidies (Hatayama, 2018). Once again, how wage subsidy programmes are designed has a significant impact on programme effectiveness, with relevant design factors including participant profiling and targeting, supervision and incentives (WWINYE, 2020d). Conditionalities around non-dismissal clauses and good information about programmes helped ensure take-up. Conditionalities also helped prevent unintended behaviours, although it was found that strict conditionalities limited take-up. Subsidies twinned with other interventions like training or job search assistance and a focus on new hiring above existing recruitment plans (to improve cost effectiveness and reduce deadweight costs) tended to have stronger results. Targeting helps improve cost effectiveness but should be used carefully, however, as too much targeting was found to increase the administrative cost and therefore discourage uptake, or stigmatise participants (Broecke, 2013; O’Higgins, 2017).

Entrepreneurship programmes

Support to young people to become entrepreneurs is increasingly common in the MENA region (Kabbani, 2019). Small enterprises make up a majority of businesses and employers in the region.^[10] Support can take the form of training, coaching or mentoring and financial support: typically seed finance in either grant or loan form. Most programmes offer a mixture of these elements in different combinations.

Positive impact on employment and income: Meta-evaluation evidence shows that entrepreneurship programmes overall have a positive impact on employment and income for youth in low- and middle-income countries, and the effects are larger when skills training is included in the programme (Kluve et al., 2017; O’Higgins, 2017). O’Higgins (2017) and S4YE (2015) suggest that of all active labour market programmes for youth (ALMPs – see Figure 1), entrepreneurship programmes have the largest impact on employment and earnings. The effect is through self-employment in new businesses – the evidence is less conclusive where programmes support existing businesses (Fox and Kaul, 2017).

WWINYE (2020b) notes that “There is not sufficient evidence to validate the causality between youth entrepreneurship promotion interventions and the creation of jobs through the newly created or expanded businesses”.^[11] A recent programme supporting entrepreneurship in Tunisia, the Souk-At-Tanmia programme (supported by the AfDB with aid funding from the UK and other donors), found that four jobs were created per business supported. However, comparison with a control group who had not received the support found similar results (Hempel, 2015).

Importance of advisory support and programme design: One very clear finding from the literature, on which all authors agree, is that the combination of advice and finance works better than either alone (S4YE, 2015; Kluve et al., 2017; IEG, 2012; Cho and Honorati, 2013; ILO, 2014; Hatayama, 2018; WWINYE, 2020b). Start-up grants are particularly important components of support – although authors disagree on whether the primary purpose/results of these projects should be to create employment (IFC, 2013), or to impact profits (Kluve et al., 2017; Cho and Honorati, 2013). The difference of opinion may reflect the importance of context: where access to credit is constrained, the inclusion of grants is critical. Where credit is not constrained, the impact of finance is less important. The impact of grants is maximised when combined with training (for example, IFC, 2013) and training may help make results more sustainable (S4YE, 2015). Programmes combining training and advisory services improved employment outcomes irrespective of whether grants were included (Kluve et al., 2017). Evidence is scarce on the effect of mentorship specifically. Training alone is not sufficient to generate significant effects for entrepreneurs (IEG, 2012). A World Bank entrepreneurship training programme in Tunisia found just a 3% increase in the probability of being self-employed and no impact from training on the overall probability of employment or earnings (IEG, 2012).

However, combined interventions are not a panacea. Although they lead to more robust results, they may not be more cost-effective than standalone programmes since the combination makes programmes costly (ILO, 2014). Separately, a meta-review found that combined programmes did not necessarily result in greater employment effects (Grimm and Paffhausen, 2014). Entrepreneurship support does not target the poorest or most vulnerable. The AfDB considers the schemes to only reach up to 3% of the unemployed: those with the “right temperament” to run a business, more highly skilled individuals and males in their 30s (AfDB, 2012).

Targeting women: The literature offers a note of caution on the job creation potential of promoting female entrepreneurship: working with SMEs is better than with micro-enterprises since the former are more likely to grow and create jobs. Because support to start-ups normally creates micro- rather than small or medium enterprises, the potential job creation effects of this support are limited (WeFi, 2019). This finding is reinforced by other authors, highlighting that support to small enterprises results in greater employment effects than support to micro-enterprises (Grimm and Paffhausen, 2014; Ismail, 2018b; Fox and Kaul, 2017). Moreover, women may face cultural barriers to engaging fully in elements of programmes, such as access to mentorship.

Entrepreneurship programmes also work in conflict: Youth entrepreneurship programmes have shown some success in situations of conflict. For example, the Youth Opportunities Programme in Uganda found that grants for training and business start-up significantly improved earnings for youth in a capital-constrained and conflict-affected region (WWINYE, 2020b). See Section 6 for more detail on the relationship between fragility and youth employment.

Sustainability – limited evidence

There is limited evidence on the long-term impact of entrepreneurship programmes as follow-up studies are rarely done (S4YE, 2015). Fox and Kaul (2017, p. 3) warn that “cash grants to finance start-up costs seem to provide a short-term boost into employment, but the effect does not seem to last long”. Kabbani (2019) notes that in MENA, youth-led start-ups are not sustainable, citing the restrictive business environment and poor access to credit as reasons for failure. The Tunisian experience is that business regulations particularly constrain the success of small firms and lead them to remain informal (World Bank, 2014). UNICEF (2018) reports that business registration costs in MENA are more than six times higher than in OECD countries. A weak entrepreneurship culture is also cited as a barrier in the MENA region (UNICEF, 2018).

Lessons learned – integrated support throughout the business lifecycle

To maximise the positive effects of entrepreneurship programmes, in addition to combining training, coaching and financial support, maintaining support through the entire business cycle and strengthening the confidence of beneficiaries as entrepreneurs are considered important (World Bank, 2014). Support over the medium-to-long term is also highlighted as important by O’Higgins (2017) and the AfDB (referring to their Souk-At-Tanmia programme in Tunisia) (AfDB, 2016). The Tunisian authorities highlight the importance of supporting entrepreneurs over the longer term. Most programmes in the country do not provide significant support once entrepreneurs have formally created their businesses, which they attribute to the high mortality rate of these enterprises (APII, 2016).

Tunisia’s Souk-At-Tanmia programme also demonstrated the need for subsequent financial support after the initial grant finance to establish businesses (AfDB, undated). This emphasises a potential unintended effect of entrepreneurship programmes: young entrepreneurs supported by grants face problems in generating working capital after start-up which can result in them taking on debt and encountering financial difficulties they would not have faced without the initial programmes of support. Careful selection of viable business plans would reduce this risk but increases the cost of programme implementation.

Because of the nuanced evidence on combinations of approaches, the success of programmes varies according to the approach taken as well as the context of the country concerned (World Bank, 2014). Some common lessons from the evidence base follow:

• Some authors note the importance of combining programmes of support to entrepreneurs with improvements to the business climate including labour regulations (such as minimum wages, job security and social insurance contributions), access to infrastructure and credit to improve impact (IFC, 2013; WeFi, 2019; AfDB, 2012).
• To maximise the job creation impact of entrepreneurship programmes, the IFC recommends focusing on the services sector in urban areas and agriculture in rural areas (IFC, 2013).
• The use of aptitude tests to select beneficiaries is recommended to improve the results of youth entrepreneurship programmes (S4YE, 2015).
• A major obstacle to entrepreneurship in MENA cited in the literature is the fragmented entrepreneurship eco-system, with a range of organisations, both public and private, working in the space but without effective coordination. For example, six ministries have a role in supporting entrepreneurship in Tunisia (OECD, 2014), thus creating a duplication of effort and inefficiencies in the system.
• Evidence from the Souk-At-Tanmia programme in Tunisia highlights the diversity of skill levels in jobs created by entrepreneurs (AfDB, undated). At programme level, specific recommendations from this study include: rewarding skilled job creation, increasing efforts to source applicants for entrepreneurship programmes in targeted regions, sharpening selection criteria to include interviews to test commitment and ability, focusing on start-ups, longer-term mentoring and coaching tailored to each beneficiary business, adding a requirement for recipients to commit their own financing to their businesses, ensuring early involvement of local banks, and rigorous monitoring and evaluation.

A review of 37 impact evaluations of entrepreneurship programmes in developing countries found that 41% included training, 67% included financial support and 21% included counselling (Cho and Honorati, 2013). This review highlights some good practices for entrepreneurship programmes:

• Private sector delivery of programmes was found to be highly correlated with programme success.
• Cash gives larger impacts than microcredit.
• Youth is correlated with programme success, but women are not associated with significant impacts other than on attitudes (thus assisting female empowerment but not employment or income).
• Access to credit has more important effects for women entrepreneurs, who face more of a constraint in accessing finance. However, the ILO (2014) notes that finance alone does not lead to increased employment.
• Evaluating programmes over the longer term is associated with improved effects.

Cho and Honorati concluded that programmes should be customised for the context (specific constraints) and the outcome of interest.

Information sharing and job matching

Information, job match or employment service interventions help young workers communicate their skills, identify vacancies and connect with employers (Kluve et al., 2017). This is typically through counselling, job search assistance and mentoring services, sometimes accompanied by job placement services or financial assistance. Evaluations of youth-focused employment services are mainly from high-income countries (HICs) and have typically been implemented by public agencies at a national scale. The evidence on employment service programmes from low- and middle-income countries is very thin (Kluve et al., 2017).

Mixed evidence on short-term effects: Although 70% of World Bank projects include support to improve information on the labour market, little is known about how accessible or helpful job information is to job seekers, particularly in low-income areas with few formal sector jobs. Studies in HICs typically found small or often non-significant effects on employment. Although one study in Germany detected positive long-term effects on youth labour market outcomes from job search assistance, changes in labour market outcomes were transitory and there was no sign of a stepping-stone effect. Fox and Kaul (2017) similarly find little evidence that job matching services are effective, and although they tend to cost less than training programmes, they are not recommended.

On the other hand, Hatayama (2018) concludes that employment matching services have improved employment and earnings. The paper notes that incentive systems for job search, such as transport subsidies, also deliver employment and income effects for rural youth. IEG (2012) notes that, of all youth employment programming, counselling and job search programmes are the types of intervention that most often achieve positive employment outcomes. However, they note that these programmes are not applicable in regions, such as MENA, where the informal sector is large. The World Bank concludes that: “In low or no growth environments in which there are no jobs, wage subsidies, skills training, and job search will matter very little to getting a job” (World Bank, 2013, p. 52).

The Bank has successfully used job search support with work-based training to reintegrate large numbers of ex-soldiers in conflict contexts (300,000 ex-soldiers including 50,000 youth, in Bosnia and Herzegovina). Young people benefited less but still saw an increase in employment of 28%, and in earnings (IEG, 2012).

Public employment services (PESs) have limited practical impact: These tend to be rarely used and the number of successful placements is limited (Kuddo, 2012). In addition, most are very poorly resourced and inadequately staffed. In Syria, Morocco and Lebanon, the ratio of registered job seekers to PES staff is high, services are usually only in urban centres and typically do not cover informal sector employers, which are often the main source of job growth (Kuddo, 2012). While access to the internet (posting of CVs and job announcements for job seekers online) in principle broadens the reach of employment services to more remote areas while lowering costs, the extent to which internet job portals have improved youth employment outcomes in developing countries – for example, reducing search time and improving the quality of job matches – and their full potential for doing so, is still not well known (S4YE, 2015).

Contextual obstacles to success – reliance on personal networks: The widespread reliance on personal networks hinders efforts to structure and professionalise job search and recruitment. Youth primarily rely on family, friends and peers for jobs (S4YE, 2015). Between 60 and 90% of respondents to a Gallup survey conducted in 2013 across the MENA region think that knowing people in high positions is critical to getting a job (IMF, 2018). Brown (2014) also highlights the importance of ‘wasta’ (political or social connections through family or friends) in the Middle East. Wasta is typically needed to secure a government job and is perceived to ‘open doors’, including for perhaps otherwise unqualified candidates.

This is inefficient: it limits the pool of potential workers available to employers and limits the number of potential positions available to the worker. Informal networks also tend to be localised, reducing the pool of firms and workers to those within a limited area. The importance (and limitations) of social networks is also highlighted by Kilimani (2017), who adds that specifically for MENA, merit-based hiring processes would support greater equality and inclusion across socio-economic strata. That said, the structural benefits of wasta have not been well studied. Early unpublished evidence suggests that wasta does serve a purpose in low-wage sectors: it mitigates information asymmetry and creates networks incentivised to perform or else risk tarnishing the reputation of not only the individual who had benefited from wasta, but also their patron.^[12]

Sustainability – inconclusive evidence and high costs per individual placed

Although the World Bank later assumed a slightly more positive stance on the short-term job creation effects of employment services (see Datta et al., 2018), its report finds no conclusive evidence of sustained impact. A review of best practice on such programmes in high-income countries concludes that employment services improved both employment and earnings among youth, but that results were moderate and the amount of evidence from developing countries is limited (WWINYE, 2020c). McKenzie looked at ten different information or job matching interventions and found these only had significant impact on employment in India (increasing employment by 2.4% over three years). Apart from the India study, the degree of sustained employment impact (that is, over 12 months) is unknown. While the cost per person involved in such programmes can be low, the cost per individual placed in a job can be substantially higher as efforts typically result in few direct instances of hiring (McKenzie, 2017).

Lessons learned – promote private sector participation and combine with skills training

A systematic review of 38 impact evaluations found that factors increasing the likelihood of success in programmes include: participation of the private sector, close monitoring and follow-up, and combining interventions like training, job search and placement support. When it comes to youth beneficiaries, combining work-based skills with smoothing the transition from school to work was most effective for employment outcomes (IEG, 2012).

Addressing information asymmetries affecting formal sector employment is unlikely to work in contexts where information asymmetries are not the primary constraint to youth employment. This includes contexts where there are few jobs in the formal sector, or where workers face mobility, skills, social or other constraints to entering labour markets (Datta, 2018). Providing enhanced access to information for rural youth, including by subsidising transport from rural areas into towns for the purposes of job search, could yield results (as in Ethiopia, for example – see Foster, 2014). Datta concludes that employment services are typically more successful when coupled with other forms of support. Addressing mobility constraints may also help remove information asymmetries (Datta, 2018).

There is disagreement on the use of private agencies to provide employment services: the AfDB (2012) finds that private agencies serve the needs of employers rather than job seekers and are likely to assist those who are easiest to place.

Direct job creation

Public employment programmes (also known as public works programmes) create labour demand where markets alone are unable to create productive employment at the necessary scale. Interventions are typically short- or medium-term and typically target unskilled, disadvantaged or long-term unemployed workers.

For youth, public employment programmes can help first-time job seekers enter the labour market and help connect unskilled or disadvantaged youth to the labour market, mitigating the risk of losing skills and other effects of long-term unemployment. These programmes sometimes go beyond stimulating demand for labour and have social protection aims – for example, helping youth during economic downturns with income support (Kluve et al., 2017). Public employment programmes are often paired with skills training or entrepreneurship support and with wage subsidies (Kluve et al., 2017).

Limited evidence of impact: Rigorous evidence of the impact and effectiveness of direct public employment programmes (often including some aspect of cash for work) in MENA is scarce. Taking a broader geographic spread, Kluve et al. (2017) found only two studies assessing employment creation effects, both of which reported zero to negative outcomes on the probability of employment after programme participation. These projects did not effectively improve youth labour market outcomes (ibid.). Kluve et al. identified one programme in Côte d’Ivoire (Emergency Youth Employment and Skills Development Project) which showed significant short-term positive impacts on probability of employment and hours worked in paid positions, but it is not clear if early gains have been sustained.

Fox and Kaul (2017, p. 36) are dismissive of public works projects, noting that they are not often evaluated “because the effects seem obvious: people have jobs for a few months’’ and may also have some displacement effects. They also note that there have been no impact evaluations that tested the hypothesis that public employment could be a good way to teach youth “employability” skills (p. 28).

Sustainability – no evidence of positive effect and some negative impact

The evidence above suggests no longer-term effect on labour market outcomes for youth. Instead, an assessment of a German job creation programme (Caliendo, 2011) found that participation had a negative impact on short- and long-term employment prospects. GIZ’s 2019 meta-review of a large number of public works interventions emphasises that they found “no robust empirical evidence that a public works program of any type generates sustainable employment over and above the public works employment itself in the medium-to-long term” (GIZ, 2019, p. 4). The same review goes on to add that while a few projects may have had some positive impact on longer-term employment, unless a specific skill enhancement aspect is added to the design, the skills gained through regular public works programmes are unlikely to enhance future employment prospects.

Employability skills/certification

While this review does not cover formal education and Technical and Vocational Education, it does extend to specific support to skills for employability. These tend to be much more focused programmes of support to youth to promote employment. Programmes can also include a focus on certification of skills including vocational skills, such as mechanics or plumbing, which can have a significant effect where large companies require suitable certification of skills to recruit these specialists.

Skills programmes may increase productivity but do not affect overall employment levels and results improve where programmes target low-income youth and women: The categorisation of employability skills in the literature varies and it is not straightforward to separate out relevant studies. Overall, training is considered to benefit employment and earning outcomes (Kluve et al., 2017). But when we focus specifically on employability skills training, findings are less positive. The IFC (2013) found mixed results from managerial and entrepreneurial training. Skills programmes yielded some increase in productivity but did not affect overall employment levels. The IFC concluded that this type of training is ineffective in the short term but that there are more positive results in the medium term, particularly for low-income youth and women. They found insufficient evidence to assess longer-term change.

In general, employability or soft skills programmes have no significant effect on employment outcomes (Datta et al., 2018; AfDB, 2012; IEG, 2012; Fox and Kaul, 2017; O’Higgins, 2017), but evidence is mixed.

Any identified positive effects are weak: a meta-review conducted by S4YE (2015) concludes that general skills training programmes increase youth employment, but only marginally, at 3%. One impact evaluation of managerial training found that sales, productivity and profits increased after the programme over and above that in the control group, but there was no impact on employment levels (IFC, 2013). The AfDB (2012) concluded that work-based learning for youth and training programmes for the unemployed are both often ineffective. They cite the weak quality and relevance of training as well as the lack of accreditation as the main reasons for this result (AfDB, 2012). The World Bank’s evaluation of youth employment programmes and the IFC’s flagship report on jobs concur with this general finding. The latter reviewed two specific impact evaluations in the area which did not find positive results on employment or business survival (IEG, 2012; IFC, 2013). Fox and Kaul (2017) also concluded that these programmes have only had limited success. Hatayama (2018, p. 2) concludes that “skills training alone might have limited positive outcomes”.

Limited evidence of impact of skills certification: There is limited evidence on the impact of skills certification programmes in the literature. One study is the World Bank’s evaluation of a programme in Chile, which was found to improve wages for those with certified skills, and not for those without. Employment became more formalised for certified beneficiaries, although this did not translate into an increase in net employment. Results were stronger for those aged over 40 (IEG, 2012). The same report notes that accreditation processes can be manipulated, which reduces the value of such programmes.

Mixed impact in MENA: In MENA, the balance of evidence does not suggest strong effects, although there are exceptions. In Egypt, Datta et al. (2018) found that beneficiaries of a training programme were more likely to be formally employed with better benefits 18 months after the training than non-beneficiaries. However, most projects had limited impact and were sometimes counterproductive. An impact evaluation of soft skills training in Jordan found that there was no impact on employment overall (Groh, 2012). The impact evaluation of Jordan’s New Opportunities for Women programme found that soft skills training did not increase the probability of employment (IFC, 2013). A Tunisian business skills programme for university students reduced formal employment because participants were more likely to become self-employed (Fox and Kaul, 2017). A large youth employment programme in Tunisia combined coaching and skills training with stipends for youth to encourage them to find jobs (World Bank, 2011). Results from this programme showed that the employability of participants decreased, leaving beneficiaries with lower employment rates than the average for Tunisian youth (World Bank, 2014).

Sustainability – insufficient evidence

There is insufficient evidence of the sustainability of impacts from skills programmes (IFC, 2013), although with broad agreement on limited employment effects overall, the issue of sustainability is less relevant than for other intervention types.

Lessons learned – involve the private sector and integrate training with other interventions

There is broad agreement on the lessons from the evidence on employability skills programmes. These include the importance of involving the private sector in training, on-the-job training, integrating training with other employment programmes and better evaluation including ex-ante vetting of suitable candidates (AfDB, 2012 and IEG, 2012; Datta et al., 2018; Kabbani, 2019).

Improving outcomes for women

The literature typically looks at a broad category of women (not specifically young women). Microfinance has some positive short-term impacts for women, but mainly for already established businesses (rather than new ventures), and while employment services generally had a low impact, this is slightly higher for women than for men (Gardiner, 2017). There is limited evidence on the effects of public employment programmes, while wage subsidies can have positive short-term impacts on women’s employment. For women, there are consistently positive impacts of skills training on their self-esteem and entrepreneurial attitudes even if there is no impact on employment or earnings outcomes, and there are sometimes also positive impacts on men’s perceptions of women’s roles and recognition of their entrepreneurial potential (Gardiner, 2017). These changes might be helpful for longer-term employment.

Microfinance and employment services show positive effects (similar to wage subsidies) but only in the short term (Gardiner, 2017). Apprenticeships are a much less positive route into employment for young women. These often reflect and reinforce strong gender stereotypes – for example, males working in carpentry, metalworking, construction and mechanics, and females concentrated in sewing/tailoring and hairdressing, with the latter typically less well paid (Fox and Kaul, 2017). Programmes that support entrepreneurship show mixed results, and successful programmes for young women tend to supplement any business skills training with mentorship and/or life skills (Fox and Kaul, 2017).

Kluve et al. found no clear answers as to what works to address youth employment for young women and no evidence that solely targeting young women will lead to better outcomes (Kluve et al., 2017). However, there is some consensus that projects focusing on broader, transferable skills supportive of positive outcomes across different sectors may be cost-effective for all young people and especially for women and girls (Fox and Kaul, 2017; S4YE, 2015; Gardiner, 2017). To capture these wider benefits, programmes need to monitor not only employment and earnings data, but also indicators for quality of life and gender equality, as well as other aspects of economic empowerment such as self-esteem, labour market participation, autonomy in life decisions, job quality, and attitudes and behaviour of men (Gardiner, 2017).

Lessons learned

The World Bank’s Adolescent Girls Initiative (AGI, 2013) draws on rigorous evidence to highlight some key factors driving impact, including strong outreach strategies to reach the most vulnerable young women, especially those not traditionally working outside the home, and the right incentives for participation and retention. Supplementary training in business and life skills can enhance participants’ resilience, as can building social and financial capital (access to savings accounts, incentives to save and invest) (World Bank, 2015).

Targeting vulnerable and marginalised youth

The lack of robust, disaggregated demographic data makes the development of effective policies aimed at marginalised and vulnerable youth particularly difficult (UN, 2018). Moreover, specific groups of young people, including for example youth with disabilities, are often not considered explicitly in data gathering and analysis (UN, 2018). It is equally important to recognise the intersectionality of disadvantage and marginalisation that young people may face, with complex layers of physical, economic and social barriers. Development actors need to beware of adopting a one-size-fits-all category that objectifies young people (Brookings, 2018).

Much of the focus on youth unemployment in MENA is on more highly educated MENA youth, perhaps masking the plight of poorer young people who cannot afford to be unemployed (the working poor). Focusing solely on unemployed youth overlooks the fact that many young people may be working but underemployed (that is, working fewer hours than desired), or with little economic gain (UN, 2018). Underemployment is difficult to measure and therefore often ignored. Young people from poor households may be forced out of education before they want to leave and do not have the option of surviving on welfare payments. They have to find work, no matter how precarious or informal (UN, 2018). The ILO highlights the sharp upward trend of extreme working poverty among young workers, particularly in the Arab states, where it increased by 12% between 1999 and 2019 (ILO, 2020b).

Addressing spatial and rural-urban dimensions of youth unemployment

In most MENA countries, more than 20% of the population is rural (in Afghanistan, Sudan, Yemen, Pakistan and Somalia, the rural population exceeds 60%). In Egypt, Iraq and Yemen, rural poverty is twice as high as urban poverty and in some rural regions of Morocco the poverty rate is 40% above the national average (IMF, 2018). Almost 60% of the poor in Algeria, Egypt, Morocco, Sudan and the OPT live in rural areas (Backeberg and Tholen, 2017).

In these rural areas and for the poor, agriculture remains the principal driver of economic development. It may be the most viable livelihood option for youth in rural areas, although young people may feel farming is ‘backward’ and be reluctant to pursue it. Stimulation of the agricultural market is important for boosting youth employment but requires innovative programming to address stigma, land scarcity or lack of land rights, lack of finance or credit, irrelevant education or training, and the inability to participate in value and supply chains (FAO, 2018).

Harnessing opportunities for youth in agribusiness entrepreneurship and innovation (including in information and communications technology) could help improve the sector’s image and productivity (Fox and Kaul, 2017; FAO, 2018; S4YE, 2015). However, there are few evaluations of programmes helping youth to develop sustainable livelihoods in farming (Fox and Kaul, 2017).

Importance of targeting and profiling

The need for careful targeting of interventions towards youth is highlighted in the literature (ILO, 2016a; Fox and Kaul, 2018; Kluve et al., 2017) and should include: identification of the main constraints faced by young people in the labour market (both demand and supply barriers), well-designed screening/profiling mechanisms to target those most in need, sound monitoring and evaluation systems to ensure that key features actually address the constraints identified, and continuous adjustment to ensure that targeting responds to changing labour demand and supply patterns (ILO, 2016a).

Kluve et al. (2017) found higher returns on investments that targeted more disadvantaged youth (cost effectiveness). Kabbani (2019) highlights targeting failures in many youth-focused employment initiatives in MENA: these are largely implemented in urban areas with limited outreach and inflexible schedules, and are thus less likely to benefit rural populations, women and less-educated, low-income groups.

There is strong evidence that programmes which profile beneficiaries and/or have systems in place to track and retain beneficiaries are more likely to succeed and have a greater effect (Kluve et al., 2017). This does not imply individualised treatment, but grouping beneficiaries into broad categories (for example from minimal support to ‘hard to serve’) seems to have a strong link to positive outcomes (Kluve et al., 2017). The Adolescent Girls Employment Initiative in Nepal (part of a World Bank pilot) incorporated a results-based system whereby training providers received different bonus payments for successfully placing participants from specific vulnerable populations in ‘gainful’ employment. The Jordan NOW project found that good targeting of disadvantaged youth was key for achieving strong outcomes (Kluve et al., 2017).

Supporting youth with disabilities

Not enough is known about the effectiveness of programmes addressing young people with disabilities (Datta, 2018). A recent ICAI review concluded that “there is little robust evidence, in any sector, about what works for disability inclusion in aid programming. This was true in a 2011 overview, and surveys since have confirmed the lack of information on disability inclusion in fields as varied as employment, education and violence against women and girls” (ICAI, 2018, p. 10).

The limited evidence available suggests that traditional vocational training programmes (specialised skills, taught separately) have not been successful in promoting employment for people with disabilities (Datta, 2018). Disabled people generally have less access to finance, including microfinance. New pilot approaches are addressing education gaps, public opinion, finding champion employers that demonstrate the productive contributions of employed youth with disabilities, and using information and communications technology to help youth access skills and training for employment. Approaches such as providing access to mainstream intermediation services and skills training, and advocacy with the public sector and with employers, appear to show promise (Datta, 2018).

In its analysis of programmes to support youth with disabilities in Syria, the Women’s Refugee Commission (WRC, 2017) emphasised the need to: 1) avoid treating young disabled people as a homogenous ‘at risk’ group; 2) shift perceptions of disability away from a sense that people with disabilities are uniformly ‘vulnerable’, taking into account varied skills, capacities and assets; 3) involve youth in programming; 4) identify intersecting factors that make youth with disabilities particularly vulnerable; and 5) pilot cash-based interventions that support transition into more sustainable livelihoods.

[8]  In Egypt, for example, the evidence showed that employment across the country’s economy slows by almost 1.5% annually when politically connected firms enter a sector where they had no previous presence (UNDP, 2013).

[9] Taxes are only slightly lower than OECD and European averages, but without the same standard of public services.

[10] For example, IEG (2012) notes that SMEs make up 80% of net job creation and 67% of employment in developing countries.

[11]  Note this does not include self-employment, but additional jobs. See Table 2, WWINYE (2020b), link.

[12]  Interview, Nader Kabbani, 23 August 2020.

Unintended consequences

One common criticism of youth employment programmes is the tendency to focus measurement of impact solely on employment and wages, with insufficient consideration given to other positive or negative effects of programmes, whether for participants, non-participants or the wider economy (Kluve et al., 2017; Gardiner, 2017; S4YE, 2015; World Bank AGI, 2013).

Positive unintended consequences: Many commentators note that focusing purely on the employment-related results (jobs and income) of interventions often fails to capture their true impact (World Bank AGI, 2013; Fox and Kaul, 2017). Contradictory findings about the efficacy of interventions may reflect the difficulty of separating skills needed for successful livelihoods from the other skills needed for successful transition to adulthood and critical decisions about education, sexual activity, family life and so on (S4YE, 2015; Fox and Kaul, 2017). Employment not only provides income, but has important effects on sense of identity, status, self-confidence, connections to others in the community and overall satisfaction with life – outcomes with tremendous value for participants and society (Fox and Kaul, 2017). Evidence suggests that combining interventions increases the chance of programme success. For example, offering services or creating integrated programmes that complement the main intervention can increase the magnitude of the effect and the probability of success by 14 to 21% (Kluve et al., 2017).^[13]

Negative unintended consequences: While some spillovers may positively affect overall employment outcomes, in certain cases they can hamper the performance of programme non-participants. S4YE (2015) raise these so-called ‘general equilibrium’ effects as important areas for future research.

Risk of entrenching poor economic management: Programmes that quickly create unnecessary jobs in already large public sectors may be difficult to wind up, can distort labour market incentives, reinforce the region’s bloated public sector (see Section 2.2), entrench social norms that privilege public sector jobs (see Section 2.2.) and may divert resources away from a potentially more vibrant private sector. Providing subsidies to educated unemployed youth could reinforce future skills mismatches if the criteria for receiving subsidies is based solely on educational credentials (WEF, 2012).

Avoiding project capture: Where youth employment programmes operate in highly politicised spaces or aim to reduce political instability, the most material risk of harm comes from project capture by vested interests. In politicised spaces, personal networks matter. Without adequately understanding this, donor-funded jobs are coveted resources and there is a material risk that programmes are ‘captured’ by powerful interests and reinforce pre-existing dynamics of clientelism and favouritism or mobilise political support (ILO, 2020). Robust programme targeting, conflict analysis and beneficiary selection are key, otherwise interventions risk exacerbating exclusion and frustration (ibid.; see also Özerdem and Podder, 2011).

Do no harm – substitution and displacement: The main unintended effects of demand-side youth employment interventions are labour substitution, that is reducing the employment of non-targeted groups in a given business as a result of creating jobs through the programme, or displacement, in other words supporting new jobs in one business at the expense of others in competing businesses.

Youth employment schemes are particularly prone to displacement effects where jobs are scarce. The risks include displacing other businesses (AfDB, 2012; Ismail, 2018b), displacing younger workers who have not benefited from being involved in a programme (particularly those from rural areas, see Section 2.3), or harming the employment prospects of older (un)employed people (Fox and Kaul, 2017). When a participant displaces a non-participant in a job, results show a positive impact at the outcome/participant level, but positive project results do not then translate to a positive aggregate change.

Programmes may be regressive if they push youth to the front of the queue at the cost of displacing qualified older or less well-off candidates (ibid.). Entrepreneurship support, for example, can displace employment in other small businesses (O’Higgins, 2017). Although substitution or displacement may not necessarily affect direct programme-level success, conflict sensitivity and ‘do no harm’ principles require that potential negative unintended consequences of programming on non-targeted groups should be monitored, understood and minimised.

Risk of beneficiary harm: Where programmes provide start-up finance to entrepreneurs, there is a potential risk of supporting businesses that are not economically viable if there is insufficient vetting of candidates and business plans. This can lead to negative outcomes for participants. In the Souk-At-Tanmia programme in Tunisia, for example, the support effectively encouraged recipients to take loans from the banking sector, using seed grants as collateral. Where businesses were not viable, the participants became indebted as a result of the programme (Hempel and Midani, 2015).

Experience from Europe suggests that reforms liberalising temporary contracts to help create additional entry options for youth may generate a fragmented labour market, that is a ‘secondary’ tier of jobs characterised by excess turnover and limited possibilities of transition from fixed-term to permanent positions (Eichhorst et al., 2016). This increases the risk of repeated spells of temporary employment and unemployment (ibid.). Another risk is that occupation-specific skills imparted by vocational training may become obsolete faster than more general problem-solving skills taught at higher education institutions. Vocational training is more likely to lead to employment in jobs that are at risk of automation and youth with fewer skills may have to switch from one precarious job to another (ILO, 2020b, p. 51).

Finally, promoting young women’s economic activity may increase the possibility of a ‘double burden’ of responsibility for unpaid domestic labour and employment. Projects also need to be aware that young female participants may experience increased violence (both within and outside the household) as they claim new space in the economic realm (World Bank AGI, 2013; S4YE, 2015).

Risk of waste: As previously explored, there is also the risk of supporting those who need less help to obtain employment: this increases costs without bringing additional benefits and leads to deadweight cost – that is the waste of subsidising jobs which would have been created anyway (Broecke, 2013; Datta et al., 2018; O’Higgins, 2017). Careful targeting may help to minimise this.

[13]  Indeed, there is early evidence that well-run programmes in communities with job growth can have positive ‘replacement effects’: if participants from sectors with excess supply are trained and placed in jobs in sectors with excess demand, they reduce competition for jobs in an over-subscribed sector (for example, the public sector) while also creating space for others in this sector.

Sustainability and cost effectiveness

Sustainability of improved employment outcomes

As discussed, there is little available evidence of the long-term effects of youth employment programming in general, and particularly in MENA (S4YE, 2015; IEG, 2012). Where programmes were evaluated, this was typically done soon after the programme ended: neither monitoring nor evaluation frameworks are typically financed beyond the end of the programme (IZA, 2014). For some government-run programmes, only activities are monitored, with no attempt to monitor impact in either the short or the long term (see OECD, 2014).

As can be seen in the preceding sections, where there is evidence, the findings on sustainability are relatively weak across the intervention typology on active labour market programmes (ALMPs). Youth employment programmes are, by definition, short-t-medium-term efforts to address weaknesses in overall labour market conditions. Raising employment, including youth employment, requires economic growth and resolving institutional and structural challenges to employment in each country, in other words the broader enabling environment activities at the top right of our typology.

Many authors highlight the importance of, for example, the business enabling environment, labour regulations (minimum wages, social insurance, job security), infrastructure availability and the growth context, as well as economic transformation, for improving employment outcomes over the longer term (AfDB, 2012a and 2012b; S4YE, 2015; Fox and Kaul, 2017; Datta et al., 2018; World Bank, 2019a; Ismail, 2018b). In MENA, labour market rigidities are common, such as complex layoff procedures and costly social insurance mechanisms which act as a disincentive to formalising employment.

ALMPs can, however, play an important role, even if the jobs they create are unlikely to be sustained. Kabbani (2019) highlights their value in managing cyclical unemployment, equally relevant to times of crisis. Moreover, the skills participants develop through their supported employment experiences can help improve, at least in theory, their future employability, although there is no evidence of this effect (S4YE, 2015).

It is important to note here that ALMPs themselves are often not sustainable due to their high cost (see Section 5.2 on cost effectiveness).

Cost effectiveness of youth employment programmes

The cost effectiveness (typically measured as cost per job created) of employment generation programmes is rarely discussed in the literature, or in programme documents, leading to a lack of conclusive evidence in this regard (IZA, 2014; Kluve et al., 2017; Fox and Kaul, 2017; Ismail, 2018b). Even where cost effectiveness is calculated, the net impact of displacement effects is not considered. Fox and Kaul (2017, p. 14) argue that, if it were considered, “many programs would be revealed as a waste of money”. That said, some data can form a useful benchmark for understanding cost effectiveness.

WWINYE (2020) notes that cost information is often underreported in employment programming. The calculation of the benefit side of cost effectiveness estimates also varies. The number of jobs created directly by an intervention is commonly reported as the benefit, but some attempts have been made to include broader employment effects across the economy (such as indirect (or suppliers’) jobs and induced jobs from additional spending by those employed by the programme). IFC (2013) conducted a meta-evaluation of 39 evaluations in the 2000s and calculated the employment multipliers achieved. They report that for each job created, up to 27 broader jobs could be assumed for some sectors and countries, while for others the multiplier was only 20% (0.2 additional jobs inferred for every job created). They report a 40% induced employment effect in Jordan.

Global benchmarks: AfDB (2015) provides an overall benchmark of USD 6,000 per job created throughout its entire portfolio across USD 1.2 billion spent between 1990 and 2010. Rates vary significantly by sector, echoing the IFC finding, with costs ranging from USD 1,600 for institutional reform efforts to USD 136,000 for health and education programming (p. 14).

In MENA, IFC (2013) calculated the average cost of direct jobs created through its private sector interventions (that is, not just youth employment but also programmes to support access to finance, infrastructure, the investment climate and skills) as ranging from USD 4,800 to USD 52,600, depending on the sector, with agriculture the most cost-effective and services the least cost-effective. In Tunisia, figures were USD 1,500 per job created in agriculture and USD 27,000 in communications.^[14] Although it is unclear how it is calculated, a business case for a DFID programme in MENA reports a benchmark estimate for employment creation programming of USD 12,000 per job created, based on an IFC methodology note about employment effects in the region (DFID, 2012).

Youth employment benchmarks: The cost effectiveness data in the evaluation of the Tunisian SIVP employment subsidy programme compares well with the global MENA benchmark: the cost per successfully placed beneficiary was USD 12,400 (World Bank, 2014). Broecke (2013) provides benchmark costs for subsidy programmes in the US as between USD 9,000 and USD 75,000 per job created, suggesting the Tunisian programme was relatively cost-effective. A regional example of a labour market matching service in Jordan cost USD 20,000 per successful job match. Although this was the only job matching estimate identified, it is probably not representative: the programme had low employment effects and therefore the few successful matches achieved were expensive (Fox and Kaul, 2017). IEG (2012) finds that job search programmes are the least expensive and some of the most successful in creating jobs for youth, implying good cost effectiveness, but it does not cite figures to evidence this point.

Low cost, high impact: Where data is provided, entrepreneurship programmes tend to cite lower costs. An entrepreneurship programme in Tunisia, Mashrou3i, had a total cost per job created of USD 3,235, although the authors note that this refers solely to programme costs and does not include the investments beneficiaries made in their own businesses (USAID, 2016). The early stage of the Souk-At-Tanmia entrepreneurship programme in Tunisia cost USD 7,400 per job created, although costs per beneficiary were much higher (since each business created an average of four jobs) (Hempel, 2015). The cost effectiveness of entrepreneurship programmes depends on how well they target youth with the right aptitude, otherwise the cost per job created can rapidly increase (S4YE, 2015).

Datta et al., (2018) find that, while average employment effects for training programmes are less than 10%, costs are high and therefore programmes tend to be cost-ineffective. This clearly depends on the approach taken to training and the employment impact achieved. They quote costs per trainee of USD 1,000 to USD 2,000, which means that each job created would cost between USD 10,000 and USD 20,000 (because only 10% of participants actually gain employment). They suggest that business training has greater cost effectiveness at USD 420 to USD 1,800 per job created, with business consulting being much less cost-effective at USD 22,800 to USD 23,900 per job created.

There is a trade-off between the scale of employability skills programmes and their cost effectiveness. Small-scale programmes were found to achieve better results but are more expensive (AfDB, 2012). Fox and Kaul (2017) cite costs of USD 1,000 to USD 2,000 per capita for life skills training.

Whether programmes lead to sustainable economic development

The literature suggests that positive employment effects of youth employment cannot be assumed a priori and that any positive effects are unlikely to be sustained. Poor design, a failure to understand and adapt to the local labour market context and short-termism are common reasons for failure. Even where jobs are created in a programme, the net effect may be low as jobs are displaced elsewhere in the economy. Programmes are often fragmented and poorly coordinated across actors or with policy contexts, meaning that even successful programmes are rarely scaled up (S4YE, 2015). As a result, it would be surprising if the programming resulted in sustained economic development. Kluve et al.’s (2017) meta-evaluation found that youth employment programmes in general do not have a statistically significant impact on business performance, or therefore on growth.

However, we have also seen some examples where significant employment returns are achieved and even sustained. The positive spillover effects on programme beneficiaries, for example of skill development, including confidence and business acumen, mean that productivity levels may well improve over the longer term and this could contribute to improved economic growth down the line.

Ultimately, the long-term impact on economic growth is difficult to ascertain, not least because it is not monitored, even by larger donors (IEG, 2012). Some programmes do not even monitor employment outcomes in the short term (Hayamata, 2018). Kabbani (2019) notes that limited evaluation of these programmes is common in the MENA region. ODI (2018) highlights the need to monitor outcomes on livelihoods resulting from programmes supporting refugee youth employment in Jordan.

The lack of wider economic impact does not necessarily mean that programmes do not add value. They have been found to provide benefits for targeted disadvantaged groups, such as youth or women, which is important both for those individuals’ livelihoods and to support the broader goals of stability and migration, discussed in Section 6.

How context affects the success of youth employment interventions

In labour economics, it is the intersection of the supply of labour and the demand for labour which matters for employment outcomes. ALMPs typically address either the supply or demand side. A common reason for failure of programmes in the past has been inadequate situating of these programmes within the local context – that is to say, supporting interventions to improve the supply of quality labour when the demand for it (in other words, jobs) was not available. In MENA, youth employment programmes need to respond to the mismatch or disequilibrium between the relatively high supply of educated youth and the low availability of formal employment due to economic stagnation or instability.

Evidential support for the importance of the economic context: The economic context of youth employment programmes matters a great deal to programme success (Hatayama, 2018; Fox and Kaul, 2017). Where there is high unemployment, ALMPs that address the supply of labour (such as skills training) will have little impact. Where there is excess labour demand, support to the demand side (such as entrepreneurship or job creation programmes) is unnecessary. Where the labour market is in equilibrium, job search and information services will not help. It is therefore of vital importance that intervention type selection as well as programme design takes adequate account of the labour market features in the local context. However, this is in aggregate, and it is important to remember that, at the margin, programmes may still assist targeted groups who are excluded from the labour market. The World Bank notes that programmes should work on both the supply and the demand side at the same time to have greatest impact (IEG, 2012). Of course, while each programme, or donor, does not need to take such a comprehensive approach, it is important that programmes are adequately situated in country contexts through the use of good diagnostic work (including labour market and political economy analysis) underpinning the theory of change and taking account of the work of other stakeholders, including government and other donors.

MENA-specific attributes: MENA economies are characterised by high levels of educated labour, a youth bulge, and low levels of formal employment (ESAY, 2017). Fox and Kaul (2017, p. ii) suggest that in these contexts “the case for supply-side youth employment interventions is in serious doubt”. Kabbani (2019) notes that most youth employment programming in MENA focuses on the supply side – technical and soft skills development – with fewer than 10% of programmes focused on the demand side.

At the same time, government programmes to generate labour demand in MENA are typically large and donor programmes should complement these efforts. For example, Tunisia spends about USD 70 million a year on ALMPs (IEG, 2012). IFC (2013) highlighted the value of using a job lens on all interventions to maximise the direct, indirect and induced employment effects of all programming in a region, so as to impact supply chains more explicitly as well as providing direct support to ALMPs.

Because of this, support to create new employment opportunities through entrepreneurship programmes may more likely help with the youth employment challenges in the region than other intervention types (Fox and Kaul, 2017). This explains their popularity in the region and in UK aid programmes. Such demand-side interventions must nevertheless take account of the broader context. ODI (2018) explains how using special economic zones in Jordan did not increase labour demand because of the perception that investment in Jordan remains risky relative to neighbouring countries. ODI also highlights the value of supporting refugees in accessing informal employment opportunities in the active informal sector of the country.

Finally, because of the fragmentation of efforts and the importance of context (both the labour market itself and the regulatory and economic context), ESAY (2017) suggests that MENA countries should develop comprehensive, integrated national employment frameworks, bringing together the multiple stakeholders supporting youth employment and addressing both the supply and the demand elements of the challenge. Donors could have a useful role engaging with policy dialogue so as to ensure a broader contextual fit for their programmes and to seek to influence the overall approach in the country.

[14] Figures are reported as jobs created per $1m investment, author’s calculations.

Fragility and youth unemployment

Evidence from available reviews shows that the link between low youth employment and conflict or extremism is weak, indirect and highly context-specific (Izzi, 2020; RUSI, 2015; Idris, 2016). Although youth employment programmes in conflict-affected countries may succeed in improving earnings for youth (WWINYE, 2020b), evaluative evidence suggests that those seeking to bolster stability or reintegration through employment (see, for example, AfDB, 2016) show no clear evidence of sustained peacebuilding outcomes (Idris, 2016, also citing WHO, 2015).

In most cases, instability makes it harder to reach youth through employment programming: the S4YE 2015 report (p. 126) finds that “not much has worked well to improve the readiness, employment and productivity of young people in fragile environments” (see also AfDB, 2016). When coupled with a limited effort to gather outcome and impact-level data, this has impeded efforts to systematically monitor the impact of employment programmes on peacebuilding (Idris, 2016).

There is a clearer link between unemployment and migration (AfDB, 2016; IZO, 2018; ILO, 2016b; UNESCO, 2017; Idris and Strachan, 2017). Migration or displacement can often be triggered by loss of livelihood due to climatic stress and/or conflict (ibid.). High rates of youth migration can also undermine growth in ‘home’ countries, compounding the growth challenge in affected nations (AfDB, 2016). Addressing this challenge, however, requires recourse to a broad range of initiatives that go beyond traditional youth employment programmes (UNESCO, 2017).

Whether youth employment reduces conflict or instability

An ILO evidence review on youth employment initiatives as a peacebuilding strategy (Izzi, 2020) notes that youth employment programmes are a common element in stabilisation strategies, but states that the evidence as to their peacebuilding impact is often assumed and “scant at best” (ibid.). The Izzi review of programmes to increase employment and reduce violence in North and sub-Saharan Africa found insufficient evidence that youth employment programmes contribute to peacebuilding, primarily because this outcome is not sufficiently monitored. The review found that projects monitor direct impacts while “assuming, rather than demonstrating, a corresponding peacebuilding impact for society at large” (p. 29).

Writing for the World Bank, Cramer (2011, p. 2) also found “no grounds empirically” for a causal connection between unemployment, violence and war. This was reinforced by Idris (2016), whose literature review of links between youth employment and violence found that programmes often assumed a causal link between unemployment and violence, without sufficient data to back this up. The Izzi review found that youth employment programmes designed to reduce instability operated on the premise that providing young people with jobs can make them more resilient to violence and promote peace (Izzi 2020). Such projects resort to stereotypes of idle young men needing to be kept busy and “out of trouble” (see also Idris, 2016). Literature reviews and project evaluations show that, while there is a correlation between unemployment and rates of political violence or conflict, this does not imply causation and does not explain why areas with similar rates of unemployment and poverty have different outcomes when it comes to stability (Izzi, 2020; RUSI, 2015; Idris, 2016).^[15]

The underlying notion that a large youth population relative to the general population, or a youth bulge (a phenomenon where the ratio of under-30s in a country surpasses the population over 30) causes violence (see ERF, 2017 or World Bank (Urdal), 2004) has been robustly challenged (Idris, 2016; Hilker and Fraser, 2009; Bricker and Foley, 2013). Although there is a correlation between relatively high youth populations and conflict risk, Bricker and Foley (2013) looked behind the correlation and found that “the causal roots of violence lie in the pressure youth cohorts exert on the total labour force, rather than the mere presence of a youth bulge itself” (p. 180).

To measure the stress that youth can exert on labour markets, Bricker and Foley introduced the ‘Youth Risk Factor’ (YRF): the ratio of the number of 17-to-26-year-olds to the size of a country’s total labour force. They found that the YRF had a statistically significant and large effect on violence and conflict. They applied this research in countries affected by protests during the Arab Spring: Tunisia, Egypt and Syria. Tunisia’s YRF value was over 0.60 in 2003 and went up to 0.63: over one and a half standard deviations above the youth risk factor mean of 0.44. This means that the Tunisian labour force was under considerable pressure to absorb young workers. Egypt’s figures are similarly pronounced: its values reached 0.72 in the late 2000s: fully two standard deviations above the YRF mean.

The Bricker and Foley data showed why the presence of large numbers of unemployed youth alone was not sufficient to explain violence. Syria had a small youth bulge (0.34, less than one standard deviation above the mean) but had “an astronomically high” YRF which reached 0.88 in 2010 (p. 190). This meant that even without a youth bulge, the number of youth was disproportionately high relative to the country’s total labour force.^[16] This data explained why Morocco (and Qatar) experienced relative calm despite their youth bulge (in other words, their total market was able to absorb the high rate of youth workers). The review reinforced the need for labour market reform to create sustainable ways for youth to be absorbed into the labour market, rather than supply-side or educational activities.

Finally, the Idris (2016) literature review reveals ambiguity about whether it is unemployment, underemployment or quality of work that is relevant, as the relative absence of robust gender and age-disaggregated data challenges any robust finding in this regard. The review concludes that although youth unemployment may contribute to violence, it is only one of many relevant social and political factors, including corruption, injustice, humiliation and dysfunctional family relationships (see also RUSI, 2017). The review cited WHO (2008) findings that multi-sectoral interventions are needed to effectively reduce youth violence.

Lessons learned: The S4YE (2015) report concluded that “not much has worked well to improve the readiness, employment and productivity of young people in fragile environments” (p. 126). Idris cited a House of Commons report and Walton (2010) to state that the lack of evaluative evidence, poor investment in research monitoring and evaluation and poor project design meant that little effort has been made to assess the impact of youth employment programmes on conflict. Izzi (2020) found that any relationship between employment and violence cannot be assumed but is complex and context-specific, thus limiting the ability to generalise and transfer findings, knowledge or lessons across contexts. For example, lessons from UK employment programmes to reduce the likelihood of youth joining gangs were found not to be relevant to attempts to stop Nigerian youth joining Boko Haram (ibid.).

Whether youth employment reduces extremism

While violent extremism or terrorism can overlap with other forms of conflict (Hilker and Fraser, 2009), this section considers the evidence of a link between youth unemployment and extremism or terrorism. According to the evaluative material consulted, any link between unemployment and extremism is “indirect, complicated and probably weak” (Krueger and Maleckova, 2002, p. 1; see also Izzi, 2020; Bhatia and Ghanem, 2017 and RUSI, 2017).

The evaluative evidence suggests that extremist organisations can build on perceptions of social and political exclusion, such as unemployment and poverty, during recruitment, but that poverty or unemployment are neither necessary nor sufficient to trigger extremism or terrorism (Idris, 2016; Rusi, 2017; Izzi, 2020). Research suggests instead that decisions to participate in political violence are highly individualised (RUSI, 2017; Izzi, 2020) and result from a complex interaction of political, social and economic factors (Idris, 2016).

The research establishes a link between education and radicalisation. Bhatia and Ghanem (2017) studied how education and unemployment affect support for violent extremism in eight Arab countries: Algeria, Egypt, Iraq, Lebanon, the OPT, Qatar, Tunisia and Yemen. Using data from the Gallup World Poll, they investigated whether educated individuals in poverty or without employment were more inclined to support violent extremism in MENA.^[17] Their research concluded that, while unemployment alone does not impact radicalisation in general, unemployment among the more highly educated leads to a greater probability of radicalisation. In this context, a little knowledge is a dangerous thing. This research for Brookings found that those educated to secondary level have the highest probability for radicalisation, followed by those with tertiary education.

The correlation between higher education levels and increased likelihood of supporting terrorism was reinforced by a 2002 study by Krueger^[18] and Maleckova. They studied the link between education, employment and radicalisation in Lebanon and Israel and found that a standard of living above the poverty line in Lebanon, coupled with secondary or higher education, was positively associated with participation in Hezbollah militant activities in the country. In Israel, Jewish settlers who attacked Palestinians in the West Bank in the early 1980s were “overwhelmingly from high-paying occupations” (p. 2). The study found that economic conditions and education are “largely unrelated to participation in, and support for, terrorism” (p.2). It cautioned against viewing terrorism as a direct response to limited market opportunities, and stated that terrorism is “more accurately viewed as a response to political conditions and long-standing feelings (either perceived or real) of indignity that have little to do with economics” (p. 3).^[19]

The findings above suggest that unemployment per se does not lead to extremism, but the mismatch between education-based expectations and opportunity can increase youth frustration. This point was reinforced in the 2016 Idris review, which found that “improvements in education can contribute to raised expectations among young people, and [create] frustration when public sector jobs were no longer available and those in the private sector were low-paid or unsuited to their skills” (Idris, 2016, p. 13).

Finally, a 2015 UK government-funded literature review by the Royal United Services Institute (RUSI) into the drivers of violent extremism studied the relationship between education, employment, poverty and radicalisation. This review challenged 17 hypotheses common to violent extremism, including whether income poverty and deprivation were major drivers of radicalisation (‘Hypothesis 13’), and whether underemployed young men “with frustrated aspirations and a limited stake in society” were particularly susceptible to radicalisation (‘Hypothesis 14’). On Hypothesis 13 (poverty), it found mixed evidence and stated that socio-economic discrimination affects terrorism recruitment but cautioned that poverty may be a side effect of some other cause and therefore “it is not possible to isolate it as a cause of violent extremism” (p. 2). The study also found mixed evidence on Hypothesis 14 (employment), with limited evidence that militant groups recruit from the ranks of the unemployed or underemployed, including well-educated unemployed youth. The report concluded that poverty, deprivation and unemployment did not sufficiently explain violent extremism but could “contribute to grievances that create a conducive environment to violent extremist groups” (p. 11). Despite this finding, the report did recommend that DFID should prioritise development interventions to stimulate job creation focused on men of fighting age (p. 45) and recommended that life skills training and youth employment programming “could help reduce vulnerability of young people” (p. 20, emphasis added).

Whether youth employment reduces irregular migration

There is more evidence to support a link between employment and migration (see, for example, IZO, 2018; ILO, 2016b; UNESCO, 2017; Idris, 2016; Idris and Strachan, 2017; AfDB, 2016). Migration is often considered a form of labour mobility due to labour market conditions and institutional ‘push’ drivers. That said, the literature is clear that how all these variables interact to trigger migration is complex, varied and highly localised (ibid.).

The link between climate change, employment and migration is key. According to UNESCO (2017), climate variability, which impacts on water availability and quality, jeopardises stability and jobs for youth, particularly in arid and semi-arid regions, resulting in migration. Migration is then triggered by the loss of livelihoods which can, in turn, result from the consequences of “failed adaptation to environmental stress”. Regional migration due to conflict can also add pressure on water resources.

The decision-making process that triggers migration differs between men and women and varies across countries. For example, livelihood loss due to water stress may both hinder and enable marriage-related migration, depending on the country in question (IZA, 2018; UNESCO, 2017). When men migrate, women take up traditional male roles but without the same rights or the ability to participate in natural resource management. The early evidence indicates that male migration reduces the left-behind women’s choices, social participation and ability to work (ibid).

Non-economic push factors that affect migration include morality, legitimacy of government and local social norms (ibid.). There is a distinction between formal and informal migration, and migration prompted by conflict (which tends to result in regional displacement), and it remains unclear what drives young people to migrate irregularly (that is, without the right visas, permissions and approvals) (IZA, 2018). Although migration is understood to be triggered by complex factors, the decision to migrate irregularly is believed to be shaped by the availability of regular migration routes, including, for example, the availability of visas, and to be weighed with factors such as differential income, a “trusted social network” in the destination country, cost of journey, risk of detention and risk of physical injury (ibid., p. 7, citing Loschmann et al., 2014). As discussed in Section 4, refugees and other irregular migrants can then face further restrictions on their availability to work when they arrive at their destination.

The IZA Institute of Labor Economics analysed data from the 2016 Sahwa Youth Survey of 15-to-25-year-olds in five MENA countries: Lebanon, Egypt, Tunisia, Algeria and Morocco. The report found that labour market drivers, specifically unemployment, job formality and social security, are important to the decision to migrate irregularly. Almost half (46%) of youth willing to migrate irregularly are unemployed and only 6% of individuals willing to migrate reported being covered by the national social security system. A third of those willing to migrate irregularly have no confidence in the legal system, while 72% believe in democracy. 70% of those willing to migrate irregularly live in urban areas.

IZA regressions show that being employed in the private sector reduces the likelihood of irregular migration by an average of 7.5 percentage points compared with those working in the public sector, and that youth covered by the national social security system are also 10 percentage points less likely to migrate irregularly compared with their peers without social security coverage. Youth are less likely to migrate when they report having a belief in the democratic ideal in their home country. Taken together, the study finds that “labour market formality acts as a firewall against irregular migration” (p. 16).

[15] Idris did, however, find some weak evidence that youth unemployment may lead to an increase in juvenile crime and/or domestic violence.

[16] Their regressions showed a correlation between higher numbers of under-14s and lower levels of conflict. According to the report, Jordan and Algeria were, at the time of writing, most at risk of instability.

[17] They concentrated on MENA because the region was the biggest supplier of foreign fighters to Iraq and Syria.

[18]  Formerly Professor of Political Economy at Princeton University, also Assistant Secretary of the Treasury for Economic Policy under US president Barack Obama.

[19] The study also identified the perception that terrorism ‘works’. Public opinion polls of young Palestinians in the OPT showed that citizens believed that armed attacks were effective and had helped them achieve rights in a way that negotiations had not. Krueger and Maleckova also cite the more developed literature on ‘hate crimes’ which, like terrorism, target individuals based on group membership rather than individual characteristics. They find that hate crimes are independent of economic deprivation and do not increase during cyclical downturns. There is mixed evidence as to whether this also applies to terrorism (in other words, whether attacks tend to increase or decrease with economic growth) and cite this as an area for future research.

Annex 1 – Key terms

Youth

There is no single accepted definition of ‘youth’ or ‘young people’. The UN, for example, has defined youth as individuals aged between 15 and 24 while at the same time defining under-18s as ‘children’ under the Convention on the Rights of the Child (UNDESA, undated). The African Development Bank’s (AfDB) definition is based on the International Conference of Labour Statisticians and extends the category to encompass 15-to-35-year-olds (AfDB, 2016).

Definitions vary even within the UK. The UK’s Youth Policy defines the cohort as aged between 13 and 19 (HM Government, 2011). Wales and Scotland both take a broader view, with Wales including those between 11 and 25 (WLGA, 2020; Youth Scotland, 2016). DFID takes a ‘lifecycle approach’ to youth, as the “time during which a young person goes through a formative transition into adulthood” (DFID, 2016, p. 3). DFID mostly considers the 10-to-24 age range but recognises that young people outside of this bracket will also be transitioning from childhood to adulthood (DFID, 2016). DFID’s definition recognises that teenagers and young adults face different challenges in their transition into work, but young migrant teenagers in the Gulf Cooperation Council, for example, might face obstacles which their older peers (also ‘youth’) do not (Silver, 2017). We take the UN definition of 15 to 24 as our central definition of youth.

Employment and transition

Certain terms are common to discussions on youth employment and we define these below. Unless otherwise stated, we refer to standard International Labour Organization definitions.^[20]

Table 2: Terminology

Quality of work

Decent work: Access to full and productive employment with rights and protections. Implies a fair income, workplace security, social protection for families, better prospects for personal development and freedom to express concerns. The Decent Work Agenda is reflected in the Social Development Goals, human rights declarations and UN resolutions, but its complex legal and social nature complicates attempts to identify and measure relevant indicators (ILO, 2020).

Vulnerable employment rate of youth: Share of young own-account workers (sometimes also referred to as self-employed) and contributing family workers in total youth employment (that is, in relatively precarious employment circumstances).

Youth working poor: Data on wages or earnings helps determine whether youth can economically support themselves from their labour or continue to live in poverty despite their wages (ILO, 2020).

Sustained or sustainable employment outcomes

Reference to ‘employment outcomes’ goes beyond reference to salaried formal jobs. The International Finance Corporation (IFC) classifies employment outcomes as direct jobs, indirect jobs (employment changes for suppliers or distributors), induced jobs (as a result of direct and indirect employees spending more to increase consumption) and secondary employment (jobs created through improved access to infrastructure). In low-income countries, jobs might more realistically be defined as “activities that generate actual or imputed income, monetary or in kind, formal or informal” (Fox and Kaul, 2017, p. 4). Youth employment programmes typically measure outcomes including the transition from unemployment to employment, an increase in hours worked, an increase in earnings per day or hour and improved employment conditions.

Sustainability goes beyond the creation of jobs to describing jobs which, at a minimum, “do not violate fundamental rights and principles at work” (AfDB, 2016, p. 11) and which are sustained over time (see, for example, van Dam et al., 2016). The IFC also notes the importance of ‘net job creation’ which addresses displacement. Job displacement occurs where programmes do not result in net job creation but recruit from the same pool of employed individuals (for example, by displacing them from competitors – see Section 4 on unintended consequences).

Sustainable economic development

The UK Aid Strategy (2015, p. 7) refers to the importance of “economic development, good governance, job creation, peace and security, and the fight against climate change” as components of “sustainable development”. Youth employment is intended to promote such sustainable economic development. The concept of sustainability developed through the 1980s refers natural system’s ability to continue providing the resources that society depends on. DFID’s own Economic Development Strategy focuses on the inclusion agenda and refers to the need for “growth that delivers for everyone” (DFID, 2017, p. 9).

[20]  Decent Work, ILO, link.

Bibliography

1. Abbott, S.G. and Stivachtis, Y.A. (2019). Demography, Migration and Security in the Middle East, E-International Relations, link.
2. Abd El-Galil, T. (2020). Arab Youth Focus on Traditional Careers, Not Emerging Ones, Report Finds, Al Fanar Media, 19 Feb 2020, link.
3. Abu-Ghaida, D. and Thacker, S. (2015). MENA Regional Synthesis Report, Workforce Development, World Bank, link.
4. Action for Children (unknown). Paying the Price. Can We Help the Most Vulnerable Young People and Avoid Unmanageable Debt?, link.
5. ADB and Plan International (2019). Youth Engagement and the SDGs: What’s the evidence?, link.
6. AfDB (2012a). Jobs, Justice and the Arab Spring: Inclusive Growth in North Africa, link.
7. AfDB (2012b). Short Term Job Creation in North Africa and the Middle East: Lessons learned from international experience, prepared for the Deauville Partnership Finance Ministers Meeting, April 2012, link.
8. AfDB (Undated). Mid-term Evaluation Report Souk At-Tanmia (pilot edition), link.
9. AfDB (2016). Jobs for Youth in Africa Strategy for Creating 25 Million Jobs and Equipping 50 Million Youth 2016-2025, link.
10. Ahmed, M. (2012). Youth Unemployment in the MENA Region: Determinants and Challenges, IMF Views and Commentaries, link.
11. Akthar, S. and Rouis, M. (2010). Economic Integration in MENA: The GCC, the Maghreb, and the Mashreq, MENA Knowledge and Learning, Quick Notes Series No 33, October 2010, link.
12. Allan, H. et al. (2015). Drivers of Violent Extremism: Hypotheses and Literature Review, RUSI, link.
13. Amirali, A. (2019). The ‘Youth Bulge’ and Political Unrest in Iraq: A Political Economy Approach. K4D Helpdesk Report, link.
14. Angel-Urdionla, D.F. et al. (2013). Building Effective Employment Programs for Unemployed Youth in the Middle East and North Africa, Directions in Development, World Bank, link.
15. APII (2017). Les problématiques de creation et de pérennisation des entreprises en Tunisie, note de synthèse, Juin, link.
16. ASDA’A BCW (2018). A Decade of Hopes and Fears. 10 Years of Arab Youth Survey, link.
17. ASDA’A BCW (2019). Arab Youth Survey 2019: A Call for Reform, White paper, link.
18. Assad, R. et al. (2018). Explaining the MENA Paradox: Rising Educational Attainment, Yet Stagnant Female Labour Force Participation, Discussion Paper Series IZA DP No. 11385, IZA Institute of Labor Economics, March 2018, link.
19. Backeberg, L. and Tholen, J. (2017). The Frustrated Generation: Youth Exclusion in Arab Mediterranean Societies, Journal of Youth Studies, Published online: 26 Oct 2017, link.
20. Barbelet, V., Hagen-Zanker, J. and Mansour-Ille, D. (2018). The Jordan Compact: lessons learnt and implications for future refugee compacts, Policy Briefing, ODI, link.
21. Barsoum, G. et al. (2015). Evaluating the Effects of Entrepreneurship Edutainment in Egypt: Randomized Controlled Trial Baseline Report, ILO Geneva, link.
22. Barsoum, G. and Sarkar, A. (2017). Youth and Employment in North Africa: A Regional Overview, ILO, link.
23. Bayt.com (2017). Fresh graduates in the MENA Survey, YouGov and Bayt.com, July 2017 – Revised, link.
24. Belkacem, L. and Mansouri. F. (2012). Global Entrepreneurship Monitor 2012 GEM Tunisia National report, link.
25. Bhatia, K. and Ghanem, H. (2017). How do Education and Unemployment Affect Support for Violent Extremism? Evidence from Eight Arab Countries, Global Economy and Development Working Paper, Brookings Institute, link.
26. Blattman, C. and Ralston, L. (2015). Generating Employment in Poor and Fragile States: Evidence from Labour Market and Entrepreneurship Programs, link.
27. Borum, R. (2011). Radicalization into Violence Extremism II: A Review of Conceptual Models and Empirical Research, Journal of Strategic Security, Vol. 4, No. 4, link.
28. Bricker, N.Q. and Foley, M.C. (2013). The Effect of Youth Demographics on Violence: The Importance of the Labour Market, International Journal of Conflict and Violence, 7(1): 179-194, link.
29. Broecke, S. (2013). Tackling graduate unemployment in North Africa through employment subsidies: A look at the SIVP programme in Tunisia, IZA Journal of Labour Policy, 2, 9, 1–19, link.
30. Brown, A.N. et al. (2015) Scoping Study: Youth Transferable Skills, 3ie Report 2015, link.
31. Brown, R.A. et al. (2014). Youth in Jordan: Transitions from Education to Employment, RAND Corporation, link.
32. Cachalia, R.C., Ndung’u, I. and Yenwong-Fai, U. (2016). Why Do Young People in Africa Join Extremist Groups? Institute for Security Studies, link.
33. Chaaban, J. (2013). Expanding Youth Opportunities in the Arab Region, UNDP, link.
34. Chen, M. and Harvey, J. (2017). The Informal Economy in Arab Nations: A Comparative Perspective, Paper for Arab Watch Report on Informal Employment in MENA Region, link.
35. Cho, Y. and Honorati, M. (2013). Entrepreneurship Programs in Developing Countries: A Meta Regression Analysis, World Bank Social Protection and Labour Discussion Paper No. 1302, link.
36. Cordesman, A.H. (2016). The Underlying Causes of Stability and Instability in MENA Region. Analytical Survey and Risk Assessment, Centre for Strategic and International Studies, link.
37. Cramer, C. (2016). Unemployment and Participation in Violence, World Development Report 2011 Background Paper, link.
38. Dadush, U. (2018). Youth Unemployment in the Middle East & North Africa, and the Moroccan case, Policy Paper 18/12, OCP Policy Centre, link.
39. Dal Santo, J., Bowling, J.M. and Harris, T.A. (2010). Effects of Work Permits on Illegal Employment Among Youth Workers: Findings of a School-based Survey on Child Labour Violations, AM J Public Health, 100(4): 635-637, link.
40. Datta, N., Elzir Assy, A., Buba, J., Johansson De Silva, S., Watson. S.M. et al. (2018). “Integrated Youth Employment Programs. A Stocktake of Evidence on What Works in Youth Employment programs”, World Bank, Washington, DC. License: Creative Commons Attribution CC BY 3.0 IGO, link.
41. David, C. et al. (2018). What Works? A Meta-Analysis of Recent Active Labour Market Programme Evaluations, Journal of the European Economic Association, Volume 16, Issue 3, link.
42. Deloitte & Touche (M.E.) (2018). Arab youth employment in the MENA region and top industries outlook, August 2018, link.
43. De Mooij, R. (2017). Getting It Right, Finance and Development, 54(1), IMF, link.
44. DFID (2012). Business case and intervention summary: Support to the World Bank Multi-Donor Trust Fund for the Middle East and North Africa Region (MENA) and the African Development Bank Multi-Donor Trust Fund for Transition in North Africa, link.
45. DFID (2013). Summary of DFID’s Regional Programme in the Middle East and North Africa 2012-2015, June 2013, link.
46. DFID (2014). Operational Plan 2011-2016 MENAD Regional Updated December 2014, link.
47. DFID (2016). Putting young people at the heart of development: The Department for International Development’s Youth Agenda, link.
48. DFID (2017). Economic Development Strategy: prosperity, poverty and meeting global challenges, link.
49. DFID (2018). DFID’s Strategy for Disability Inclusive Development 2018-23, link.
50. Dibeh, G., Fakih, A. and Marrouch, W. (2017). Decision to Emigrate Amongst the Youth in Lebanon, International Migration, 56:1, link.
51. Dibeh, G., Fakih, A. and Marrouch, W. (2018). Labour Market and Institutional Drivers of Youth Irregular Migration: Evidence from the MENA Region, Discussion Paper Series, IZA Institute of Labor Economics, link.
52. Dimova, R., Elder, S. and Stephan, K. (2016). Labour Market Transitions of Young Women and Men in MENA, ILO, link.
53. Doing Business database (2020) Ease of Doing Business Rankings. Accessed 30 June 2020, link.
54. Drine, I. (2012). Youth Unemployment in the Arab World: What Do We know? What is the Way Forward? UNU-Wider Blog, June 2012, link.
55. Eichhorst, W. and Rinne, U. (2016). Working Paper Promoting Youth Employment in Europe: Evidence Based Policy Lessons, IZA Policy Paper, No. 119, link.
56. El Feki, S., Heilman, B. and Barker, G., Eds. (2017). Understanding Masculinities: Results from the International Men and Gender Equality Survey (IMAGES) – Middle East and North Africa. Cairo and Washington, D.C.: UN Women and Promundo-US, link.
57. El Kadi, T.H. (2019). Youth Unemployment: North Africa’s Ticking Time Bomb, Atlantic Community, link.
58. ESAY (2017). Youth Employment in the Middle East and North Africa. Translating Research Into Scaled Up Action: Evidence Symposium on Adolescent and Youth in MENA, November 21-22 2017, Amman, Jordan, Background Paper, link.
59. FAO (2018). Youth Employment in Agriculture as a Solid Solution to Ending Hunger and Poverty in Africa: Engaging through Information and Communication Technologies (ICTs) and Entrepreneurship, link.
60. Farasin, F., Battaloglu, C. and Bensaid, A.A. (2017). What is Causing Radicalism in the MENA? Arab Centre for Research and Policy Studies, link.
61. Fardoust, S. (2016). Economic Integration in the Middle East, Prospects for Development and Stability, Middle East Institute Regional Cooperation Series, June 2016, link.
62. Fox, L. and Kaul, U. (2017). The evidence is in: How should youth employment programs in low-income countries be designed?, USAID, link.
63. Gardiner, D. (2017). Promoting Women’s Empowerment in the Middle East and North Africa: A Rapid Evidence Assessment of Labour Market Interventions, IFAD Impact Brief, Issue 9, link.
64. Georgia, T. (2015). DFID Guidance Note on Addressing Violence Against Women and Girls, link.
65. GIZ (2019). Design and Implementation Features of Public Works Programmes, What Do We Know About Drivers of Success and Failure to Achieve Programme Objectives?, link.
66. Glenerster, R. et al. (2018). A Practical Guide to Measuring Women’s and Girls’ Empowerment in Impact Evaluations, J-PAL 2, Abdul Latif Jameel Poverty Action Lab, link.
67. Gouda, M. and Marktanner, M. (2017). Muslim Youth Unemployment and Expat Jihadism – Bored to Death? Economic Research Forum Working Paper Series, link.
68. Grimm, M. and Paffhausen, L. (2014). Do Interventions Targeted at Micro-Entrepreneurs and Small and Medium-Sized Firms Create Jobs? A Systematic Review of the Evidence for Low and Middle Income Countries, IZA DP No. 8193, May 2014, link.
69. Groh, M., McKenzie, D., Shammout, N. and Vishwanath, T. (2014). Testing the importance of search frictions, matching, and reservation prestige through randomized experiments in Jordan, World Bank – Impact Evaluation series. Policy Research Working Paper No. WPS 7030, link.
70. Groh, M., Krishnan, N., McKenzie, D. and Vishwanath, T. (2012). Soft skills or hard cash? The impact of training and wage subsidy programs on female youth employment in Jordan, World Bank, Policy Research Working Paper Series No. 6141, link.
71. Hagen-Zanker, J. et al. (2017). Cash Transfers for Refugees. The Economic and Social Effects of a Programme in Jordan, ODI, link.
72. Haidar, H. (2016). Barriers to youth work opportunities (K4D Helpdesk Research Report), Birmingham, UK: GSDRC, University of Birmingham, link.
73. Hammer J. and Pivo, G. (2016). The Triple Bottom Line and Sustainable Economic Development Theory and Practice, Economic Development Quarterly, Volume: 31 issue: 1, page(s): 25-36, link.
74. Hatayama, M. (2018). Youth employment programmes: What works in rural settings? K4D Helpdesk Report 468. Brighton, UK: Institute of Development Studies, link.
75. Hempel, K. and Midani, M. (2015). Souk At-tanmia: Independent evaluation of the 2012-2013 pilot edition in Tunisia, link.
76. HMG (2011) Positive for Youth: A new approach to cross-government policy for young people aged 13 to 19, link.
77. HMG (2015). UK aid: tackling global challenges in the national interest, link.
78. Ibarrarán, P., Kluve, J. and Ripani, L. (2018). Experimental Evidence on the Long-Term Effects of a Youth Training Program, Cornell University ILR School, link.

1. Idris, I. (2016). Youth Unemployment and Violence Rapid Literature Review, GSDRC UK, link.

1. Idris, I. and Strachan, A. (2017). Economic Drivers of Conflict in the Western Balkans, K4D Helpdesk Report, link.
2. International Crisis Group (2016). Fight or Flight: The Desperate Plight of Iraq’s “Generation 2000”, Middle East Report No. 169, link.
3. IEG (2012). World Bank and IFC Support for Youth Employment Programs, Washington, D.C.: World Bank, link.
4. IFAD (2019). Action Plan for Rural Youth 2019-2021, link.
5. IFC (2013). IFC Jobs Study: Assessing Private Sector Contributions to Job Creation and Poverty Reduction, link.
6. IFC (2018). Educating Students for Jobs, Stability and Growth. Luminus: Transforming Vocational Education in Jordan, link.
7. IFC (2019). Private Sector and Refugees. Pathways to Scale, link.
8. ILO (2014). Effectiveness of Entrepreneurship Development Interventions for women entrepreneurs: An ILO-WED Issue Brief, link.
9. ILO (2016a). Design and Targeting of Youth Employment Programmes: What Works? Presentation, Riga, 12-13 April, link.
10. ILO (2016b). SME Promotion Agencies: Is There a Best Set-up? A Quest for Good Practices, link.
11. ILO (2017a). The Global Initiative on Decent Jobs for Youth, The Global Initiative for Action, link.
12. ILO (2017b). Promoting Youth Employment and Empowerment of Young Women in Jordan. An Assessment of Active Labour Market Policies, link.
13. ILO (2017c). Towards Policies Tackling the Current Youth Employment Challenges in Eastern Europe and Central Asia, link.
14. ILO (2017d). A Challenging Market Becomes More Challenging: Jordanian Workers, Migrant Workers and Refugees in the Jordanian Labour Market, link.
15. ILO (2018a). Forum on Evidence and Impact Measurement Event Report: What Works in Youth and Women’s Employment in MENA? (Event supported by IFAD and CIDA took place in Cairo, June 2018), link.
16. ILO (2018b). Guide on Measuring Decent Jobs for Youth: Monitoring, Evaluation and Learning in Labour Market Programmes, link.
17. ILO (2018c). Constraints and Good Practice in Women’s Entrepreneurship in MENA, Impact Report Series Issue 10, link.
18. ILO (2020a). Global youth employment trends for Youth 2020: Technology and the future of jobs, Geneva: ILO, 2020, link.
19. ILO (2020b). Global Employment Trends for Youth 2020: Arab States, Briefing Note, 9 March 2020, link.

99. ILO (2020c). ILOSTAT Database. Data retrieved 1 March 2020, accessed 30 June 2020, link.
100. ILO (2020d). ILOSTAT Database. Data retrieved 1 March 2020, accessed 30 June 2020, link.
101. ILO (unknown). Jobs Strategy. Promoting Transformation and Job Creation in the Maldives, link.
102. ILOSTAT Explorer (2019). Share of Youth Not in Employment, Education or Training (NEET) by Sex — ILO modelled estimates, November 2019 (%) – Annual. Accessed 30 June 2020, link.
103. ILO-SIDA (2019). Young People not in Employment, Education or Training (NEET), Technical Brief #3, link.
104. IMF (2014). Towards New Horizons. Arab Economic Transformation Amid Political Transitions, link.
105. IMF (2018). Opportunity For All Promoting Growth And Inclusiveness In The Middle East And North Africa, Departmental Paper No. 18/11, link.
106. IOM (2016). Migration to, from and in the Middle East and North Africa, Data Snapshot, link.
107. IOM (2019). World Migration Report 2020, link.
108. Ismail, Z. (2018a). The Impact of Entrepreneurship Training Programmes. K4D Helpdesk Report. Birmingham UK: University of Birmingham, link.
109. Ismail, Z. (2018b). Lessons learned from youth employment programmes in developing countries, K4D Helpdesk Report. Birmingham, UK: University of Birmingham, link.
110. Izzi, V. (2020). Promoting Decent Employment for African Youth as a Peacebuilding Strategy, Evidence Synthesis Paper Series, ILO, link.
111. J-PAL. (2013). J-PAL Youth Initiative Review Paper, Abdul Latif Jameel Poverty Action Lab, link.
112. Kabbani, N. (2019). Youth employment in the Middle East and North Africa: Revisiting and Reframing the Challenge, Brookings Doha Centre Policy Briefing, February 2019, link.
113. Kautz, T. (2014). Fostering and Measuring Skills: Improving Cognitive and Non-Cognitive Skills to Promote Lifetime Success, OECD Paris, link.
114. Kherfi, S. et al. (2018). Labour Market Challenges of Palestinian Refugees in Lebanon. A Qualitative Assessment of Employment Service Centres, ILO, link.
115. Kilimani, N. (2017). Youth Employment in Developing Economies: Evidence on Policies and Interventions, IDS Bulletin, Vol. 48, No. 3, 2017, link.
116. Kluve, J., Puerto, S., Robalino, D., Romero, J.M., Rother, F., Stoterau, J., Weidenkaff, F. and Witte, M. (2017). Interventions to Improve Labour Market Outcomes of Youth: A Systematic Review of Training, Entrepreneurship Promotion, Employment Services and Subsidized Employment Interventions, Campbell Collaboration. International Development Coordinating Group, link.
117. Krishnan, N. et al. (2016). Uneven Odds, Unequal Outcomes. Inequality of Opportunity in the Middle East and North Africa, World Bank, link.
118. Krueger, A.B. and Maleckova, J. (2002). Education, Poverty, Political Violence and Terrorism: Is There a Causal Connection? National Bureau of Economic Research Working Paper Series, link.
119. Lippman, L.H. et al. (2015). Key “Soft Skills” that Foster Youth Workplace Success: Towards A Consensus across Fields, Child Trends, USAID.FHI 360 USA, link.
120. Liu, E. and Roth, B.N. (2017). Keeping the Little Guy Down: A Debt Trap for Lending with Limited Pledgeability, MIT Economics, link.
121. Mansour-Ille, D. and Samman, E. (2017). The Future of Work for Syrian Refugee Women in Jordan: Is the “Gig Economy” an Opportunity?, link.
122. Massoud, K. and Nadereh, C. (2016). Women, Work, and Welfare in the Middle East and North Africa, Imperial College Press, London, link.
123. McLean-Hilker, L. and Fraser, E. (2009). Youth Exclusion, Violence, Conflict and Fragile States, SDD UK Report prepared for DFID Equity and Rights Team, link.
124. McKee, M. et al. (2017). Demographic and Economic Material Factors in the MENA Region, Working Paper No. 3, MENARA (EU Funded), link.
125. MercyCorps (2020). Youth and Consequences. Unemployment, Injustice and Violence, link.
126. Miletto, M. et al. (2017). Migration and Its Interdependencies with Water Scarcity, Gender and Youth Employment, UNESCO, link.
127. Neaime, S. (2019) Recent Financial and Debt Crisis: Is the MENA Region Immune? The ERF Policy Portal Forum, 1 April 2019, link.
128. OECD (2013). Local Strategies for Youth Employment. Learning from Practice, link.
129. OECD (2014). Supporting entrepreneurship in the vocational training system in Tunisia, OECD Reviews on Skills and Competences for Entrepreneurship, link.
130. OECD (2016a). Strengthening governance and competitiveness in the MENA region for stronger and more inclusive growth, Better Policies, OECD Publishing, Paris, link.
131. OECD (2016b). Youth in the MENA Region: How to Bring Them In, OECD Publishing, Paris, link.
132. OECD (2018). Enhancing SME Access to Diversified Financing Instruments. Plenary Session 2, Discussion Paper, SME Ministerial Conference, link.
133. OECD (2020). DAC List of ODA recipients – Effective for Reporting on 2020 Flows, accessed 30 June 2020, link.
134. O’Higgins, N. (2017). Rising to the youth employment challenge: New evidence on key policy issues, International Labour Office – Geneva: ILO, 2017, link.
135. Özlap, O.N. (2011). Where is the Middle East? The Definition and Classification Problem of the Middle East as a Regional Subsystem in International Relations, TJP Turkish Journal of Politics 2:2, Winter 2011, pp. 5-21, link.
136. Paasonen, K. and Urdal, H. (2016). Youth Bulges, Exclusion and Instability: The Role of Youth in the Arab Spring, Conflict Trends, 3. Oslo: PRIO, link.
137. Paniagua, G. and Denisova, A. (2012). Meta-evaluation on job creation effects of private sector interventions, World Bank, link.
138. Population Reference Bureau (2009). Youth Unemployment and Underemployment in Africa Brings Uncertainty and Opportunity, link.
139. Premand, P., Brodmann, S., Almeida, R., Grun, R. and Barouni, M. (2011). “Entrepreneurship Training and Self Employment among University Graduates: Evidence from a Randomized Trial in Tunisia”, link.
140. Price, R. (2019). Youth Employment Needs in Nigeria, K4D Helpdesk Report, Institute of Development Studies, link.
141. Purfield, C. et al. (2018). Opportunity for All : Promoting Growth and Inclusiveness in MENA, IMF Departmental Paper, 18/11. Washington, DC: International Monetary Fund, link.
142. Rankin, K. et al. (2015). Youth and Transferable Skills : an evidence gap map, 3ie, link.
143. Raz, D. (2019). Arab Barometer: Youth in Middle East and North Africa, Arab Barometer Wave V 2018-2019, August 2019, link.
144. Rezk, M. (2016). The Political Economy Violence in Egypt, IDS Bulletin, link.
145. Richards, A. (1995). Economic Roots of Instability in the Middle East, Middle East Policy Council, link.
146. Rocha, R. et al. (2011). Financial Access and Stability: A Road Map for the Middle East and North Africa (English), MENA Development Report, Washington, DC: World Bank, link.
147. Roy, O. (2015). What is the Driving Force behind Jihadist Terrorism? – A Scientific Perspective on the Causes/Circumstances of Joining the Scene, Herbst-Tagung Autumn Conference 2015, link.
148. Rubin, M. and Katulis, B. (2019). Seven Pillars. What Really Causes Instability in the Middle East? Washington Institute, link.
149. Salehi-Isfahani, D. (2012). Education, Jobs and Equity in MENA, Comparative Economic Studies, 54(4), pp. 843-861, link.
150. Schiffbauer, M. et al. (2014). Jobs or Privileges: Unleashing the Employment Potential of the Middle East and North Africa (English), MENA Development Report, Washington, DC: World Bank, link.
151. Serageldin, M. et al. (2015). Urban Migration Trends in the Middle East and North Africa Region and the Challenge of Conflict-Induced Displacement, IOM, link.
152. Setti, K. (2017). Entrepreneurial intentions among youth in MENA countries: effects of gender, education, occupation and income, International Journal of Entrepreneurship and Small Business, link.
153. Silatech (2017). Youth Unemployment, Violence Extremism and Job Creation for Peace and Security – SILATECH as a Case Study, SILATECH, link.
154. Silver, H. (2017). Dimensions of Youth Exclusion in the Middle East, The Muslim world, January 2017, pp. 13-40, link.
155. Simpasa, A.M., Shimeles, A. and Salami, A.O. (2015). Employment Effects of Multilateral Development Bank Support: The Case of the African Development Bank, African Development Bank Group Working Paper Series No. 221 – April 2015, link.
156. S4YE (Goldin, N. and Hobson, M. with Glick, P., Lundberg, M. and Puerto, S.) (2015). “Toward Solutions for Youth Employment: A Baseline for 2015.” Solutions for Youth Employment, Washington D.C., link.
157. Toli, K. (2019). The Water-Employment-Migration Nexus, link.
158. Torelli, S.M. (2017). Explaining the Recent Spike in Migration from Tunisia to Europe, Commentary, link.
159. UK Commission for Employment and Skills (unknown). Precarious Futures? Youth Employment in an International Context, link.
160. UN (2011). Youth Participation in Development Summary Guidelines for Development Partners, link.
161. UNCTAD (2019). Chapter IV, Special Economic Zones, link.
162. UNDESA (unknown). Frequently Asked Questions, link.
163. UNDESA (2018).The World Youth Report: Youth and the 2030 Agenda for Sustainable Development, link.
164. UNDP (2016). Human Development Report 2016: Youth and the Prospects for Development in a Changing Reality, link.
165. UNESCWA (2015). Inclusive Social Development, link.
166. UNICEF (2018). Solutions that work: Evidence Symposium On Adolescents And Youth In MENA, Second Edition, September 25-26, 2018 Summary Report, link.
167. UNICEF (2019a). MENA Generation 2030: Investing in Children and Youth Today to Secure a Prosperous Region Tomorrow, April 2019, link.
168. UNICEF (2019b). Middle East and North Africa, Adolescent Development and Participation (ADAP) Section (2019). Solutions That Work: 2nd Evidence Symposium on Adolescents and Youth in the Middle East and North Africa, 25-26 September, 2018 – Summary Report, Amman, Jordan, link.
169. UNRISD (2016). Redrawing the Welfare Map: New Directions in Social Policy in the MENA Region Project Brief: 15, link.
170. Urdal, H. (2004). The Devil in the Demographics: The Effect of Youth Bulges on Domestic Armed Conflict, 1950-2000. Social Development Papers. Conflict Prevention and Reconstruction Series, No. CPR 14, Washington DC: World Bank, link.
171. USAID (2016). Evaluation: Final Performance Evaluation of “Tackling Youth Employment in Tunisia”, Mashrou3i, link.
172. Van Dam, K., Van Vuuren, T. and Kemps, S. (2016). Sustainable employment: the importance of intrinsically valuable work and an age-supportive climate, the International Journal of Human Resource Management. 28. 1-24, link.
173. Veiga, M.G. (2019). The Financing of Small and Medium-Sized Enterprises: An Analysis of the Financing Gap in Brazil, European Business Organization Law Review, 20, 633-664, link.
174. WeFi (2019). Women’s entrepreneurs finance initiative (We-Fi) Annual Report 2017 • 2018, link.
175. Welsh Local Government Association (2020). Youth service, link.
176. Werner, E. and Ulf, R. (2016). Promoting Youth Employment in Europe: Evidence-based Policy Lessons, IZA Policy Paper No. 119, link.
177. What Works Centre for Local Economic Growth (2016). Access to Finance. Evidence Review 4, link.
178. Woman’s Refugee Commission (2017). Vulnerability- and resilience-based approaches in response to the Syrian crisis: Implications for women, children and youth with disabilities, link.
179. World Bank (2011). The AMAL Program: “Hope” For Unemployed Youth in the MENA Region. MNA Knowledge And Learning Fast Brief No. 88, link.
180. World Bank (2014). Breaking the Barriers to Youth Inclusion, link.
181. World Bank (2015). The Spirit of Boldness: Lessons from The World Bank’s Adolescent Girls Initiative, link.
182. World Bank (2016). Middle East and North Africa. Public Employment and Governance in MENA, link.
183. World Bank (2017a). The Economics of Post-Conflict Reconstruction in MENA. MENA Economic Monitor, link.
184. World Bank (2017b). Iraq, Systematic Country Diagnostic, link.
185. World Bank (2018a). Poverty and Shared Prosperity 2018 : Piecing Together the Poverty Puzzle. Washington, DC: World Bank, link.
186. World Bank (2018b). Population Statistics, link.
187. World Bank (2019a). World Development Report 2019: The Changing Nature of Work, Washington, DC: World Bank. Doi:10.1596/978-1-4648-1328-3. License: Creative Commons Attribution CC BY 3.0 IGO, link.
188. World Bank (2019b). Creating Markets in Morocco. Country Private Sector Diagnostic, link.
189. World Bank (2020a). Middle East and North Africa: Overview. Accessed on 26 May 2020, link.
190. World Bank (2020b). Women, Business and the Law, link.
191. World Bank (2020c). Unemployment Youth Total (% of Total Labour Force Ages 15-24) (Modelled ILO Estimate), accessed 18 June 2020, link.
192. World Bank national accounts data, and OECD National Accounts data files (2020). GDP growth (annual %) (NY.GDP.MKTP.KD.ZG). Accessed on: 30 June 2020, link.
193. World Economic Forum (2012). Addressing the 100 Million Youth Challenge Perspectives on Youth Employment in the Arab World in 2012, link.
194. World Economic Forum (2017). The Future of Jobs and Skills in the Middle East and North Africa: Preparing the Region for the Fourth Industrial Revolution, Executive Briefing, May 2017, link.
195. World Economic Forum (2019). Global Gender Gap Report 2020, Cologny/Geneva, ISBN-13: 978-2-940631-03-2, link.
196. WWINYE (2020). What works in youth employment The Evaluation Evidence, website, accessed 22 May 2020, link.
197. WWINYE (2020b). What works in youth employment Evidence Entrepreneurship Programmes, website, accessed 23 May 2020, link.
198. WWINYE (2020c). What works in youth employment Evidence Employment Services Programmes, website, accessed 23 May 2020, link.
199. WWINYE (2020d). What works in youth employment Evidence Subsidized Employment Programmes, website, accessed 23 May 2020, link.
200. Youth Development Portal (2017). Youth Employment in MENA, Background Paper, link.
201. Youth Scotland (2016). Strategy for 2016-2019: Changing lives through youth work, link.