83925 DISCUSSION PAPER NO. 1308 Youth Employment A Human Development Agenda for the Next Decade David Robalino, David Margolis, Friederike Rother, David Newhouse and Mattias Lundberg June 2013 Youth Employment: A Human Development Agenda for the Next Decade David Robalino, David Margolis, Friederike Rother, David Newhouse, Mattias Lundberg1 June 2013 ABSTRACT This paper reviews the main challenges facing countries in attempting to improve labor market outcomes among youth, focusing on the issues that became starkly visible during the recent financial crisis. In order to better identify and set up human development interventions, the paper proposes an agenda that focuses on three areas: (1) improving the understanding of the causes and consequences of poor labor market outcomes for youth; (2) continuing to learn from the evaluation of pilots and programs that aim to promote productive employment among young people; and (3) addressing implementation issues which frequently overwhelm the best designs. The paper utilizes research on youth employment to take stock of youth labor market outcomes across regions, focusing on inactivity, unemployment, and employment indicators. A review of what is known about current interventions, including those that appear in the Youth Employment Inventory database of programs, provides the basis for determining the efficacy of five categories of intervention: (i) skills training (including vocational training, on-the-job-training programs, literacy and numeracy programs, second-chance and equivalency programs, and soft-skills programs); (ii) entrepreneurship promotion (financial assistance, technical assistance, and entrepreneurship training); (iii) subsidized employment (including wage subsidy programs, public works, and public/community service programs); (iv) employment services (including search assistance and access to labor market information, job counseling and placement services, and financial assistance for job search); and (v) reforms to labor market regulation (including anti-discrimination legislation) training programs, wage subsidies. Finally, the 1 The team thanks Jean Lee and Afia Tasneem for their support for this project. 1 paper proposes an agenda for research and policy analysis in the area of human development that is expected to help both deepen the understanding of youth employment issues and improve the selection, design, and implementation of youth employment programs. JEL Classification: F22, H55 Keywords: youth, employment, skills training, youth employment inventory, entrepreneurship, jobs, labor markets, wage subsidies 2 ABSTRACT .............................................................................................................................. 1 1. Introduction ....................................................................................................................... 5 2. Youth Labor Market Outcomes and Social Welfare ......................................................... 6 Figure 1: Expected Increase in the Youth Population by Region.................................................... 7 Figure 2: Share of Youth in the Population and Labor Force ......................................................... 8 Figure 3: Level of Education among Youth Participating in the Labor Force ................................. 9 Figure 4: Adult vs. Youth Unemployment .................................................................................... 10 Figure 5: Youth Activity Status by Region..................................................................................... 10 Figure 6: Youth Employment Status across Regions .................................................................... 13 Figure 7: Labor Transitions among Youth in Brazil and Mexico ................................................... 13 3. A Framework for Action on Youth Employment ............................................................ 14 Table 1: Constraints to Youth Employment ................................................................................. 16 3.1 Individual Characteristics ..................................................................................................... 16 3.2 Market and Government Failures ........................................................................................ 19 3.3 Macroeconomic, Demographic and Geopolitical Factors .................................................... 22 Figure 8: Job Creation and Destruction Rates in the United States ............................................. 23 3.4 Policies and Programs .......................................................................................................... 24 Table 2: Policy and Program Responses to Constraints to Youth Employment ........................... 26 4. What Have Countries Been Doing? What Has Worked? ................................................ 27 Figure 9: Youth Employment Policies and Programs, by Region and Type of Intervention ......... 28 4.1 Training Programs ................................................................................................................ 29 4.2 Entrepreneurship Promotion Programs ............................................................................... 32 4.3 Employment Services ........................................................................................................... 35 4.4 Subsidized Employment Programs ....................................................................................... 38 4.5 Labor Market Reform ........................................................................................................... 43 4.6 A Summary of Labor Market Programs for Youth ................................................................ 44 Table 3: Anticipated Effects of Different Programs by Setting and Beneficiary Type .................. 44 5. An Agenda Going Forward ............................................................................................. 45 5.1 Diagnosis –Understanding the Causes and Consequences of Youth Unemployment ......... 45 5.2 Treatment – Designing and Implementing Effective Interventions ..................................... 47 5.3 Monitoring, Evaluation, and Learning .................................................................................. 49 3 Table 4: Impact Evaluations of Youth Employment Programs (from the YEI).............................. 49 6. Conclusions ..................................................................................................................... 50 4 1. Introduction This paper reviews the main challenges facing countries in attempting to improve labor market outcomes among youth and proposes an agenda to guide future work in the area of human development. As discussed in more detail below, many young people around the world face disappointing prospects when it comes to jobs. In most countries, unemployment rates among youth are much higher than for adults and those working are often employed in precarious activities that provide little in terms of income and security. Since the financial crisis, the share of youth neither participating in the labor force nor studying has been increasing. Governments have tried to address these problems through different types of intervention, ranging from training and employment services to wage subsidies and programs that support entrepreneurship in rural areas. There have been a number of positive experiences, and our knowledge of the types of interventions that seem to work better has improved tremendously. However there are still many open questions both in terms of the causes behind poor labor market outcomes among youth and the best ways to address them. The paper develops a comprehensive framework to understand youth employment issues and guide the design of interventions to address them. It argues that several of the policies that are needed to improve labor market outcomes among workers in general will also contribute to addressing youth employment problems. These include policies that provide incentives to create new businesses and expand current ones and interventions to promote rural development. In addition, policymakers, particularly during the short-term, will also need to address market and government failures that constrain successful transitions into wage and self-employment; such failures can often be even more limiting for youth than individual limitations. These market and government failures often stem from skills mismatches, lack of information about jobs and business opportunities, lack of capital, or regulations that discourage formal labor contracts. These problems can have a greater effect on young people for a number of reasons: because they have less experience working or searching for jobs, they are less connected to jobs through their social networks, and 5 they are likely to be more credit constrained. Human development policies have an important role to play in all of these areas. In order to better identify and set up human development interventions, the paper proposes an agenda that focuses on three areas: (1) improving the understanding of the causes and consequences of poor labor market outcomes for youth; (2) continuing to learn from the evaluation of pilots and programs that aim to promote productive employment among young people; and (3) addressing implementation issues which frequently overwhelm the best designs. These issues include the identification of effective interventions; targeting, registration, and identification of beneficiaries; enforcement of conditionalities; and contracting and payment systems for providers. The paper is organized in three sections. The next section takes stock of youth labor market outcomes across regions, focusing on inactivity, unemployment, and employment indicators. The two sections that follow propose a conceptual framework to guide the design of programs to improve labor market outcomes among youth and review what is known about current interventions. The last section discusses an agenda for research and policy analysis. 2. Youth Labor Market Outcomes and Social Welfare Young people today make up about one-quarter of the world’s population, with the majority of those young people living in Asia. Countries in Sub-Saharan Africa (SSA), the Middle East and North Africa (MENA), and South Asia (SAR) are in the midst of a demographic transition that will rapidly increase the share of youth in the total population and the labor force.2 In Sub-Saharan Africa, for instance, the number of youth will continue to grow rapidly, increasing from 174 million in 2010 to 269 million in 2030 (see Figure 1). In the Middle East and North Africa as well as South Asia the youth population will also increase by around 20 percent over the next two decades. Across countries in these regions, youth will represent over 40 percent of the population and over 30 percent of the labor 2 See Mason, A. (2005). 6 force (see Figure 2). In contrast, in Eastern Europe and Central Asia (ECA) and in East Asia (EAP) the youth population will contract, while it will remain stable in Latin America (LAC). Figure 1: Expected Increase in the Youth Population by Region Source: World Bank Staff, based on data from HNPstats For those regions with a growing number of young people, the challenges in terms of job creation are considerable. Other things being equal, sub-Saharan economies will need to create around 4.6 million jobs per year to absorb the projected increase in the youth population – a 4.2 percent increase in total youth employment each year. In the Middle East and North Africa region as well as South Asia, employment would need to expand by 0.4 and 1.2 percent per year respectively just to employ young workers entering the labor market. If this expansion in employment takes place, youth will make a major contribution to increasing standards of living in the countries where they live. In Africa, the demographic transition has started particularly late but is expected to pick up steam in the coming decades, as the ratio of working age to total population increases from about 1 to 2 over the next 4 decades. This demographic dividend has the potential to contribute 1.5 to 2 percentage points to GDP growth per year, as it did in East Asia and Latin America between 1970 and 2000.3 Unfortunately, for many youth, work prospects remain slim. 3 See Mason, A. (2005). 7 Figure 2: Share of Youth in the Population and Labor Force Source: World Bank Staff, based on International Income Distribution Database Even in regions where the population of youth is not expected to expand, there are concerns regarding current labor market outcomes. Youth are often more likely to be unemployed than adults, a majority have low levels of education, many end-up in low productivity jobs (often in the informal sector), and others have withdrawn from the labor force and become idle. The low level of education of the youth population is an important part of the problem. With the exception of the East Asia and Pacific region, less than 10 percent of youth have post-secondary education. In the Middle East and North Africa and in Sub-Saharan Africa over 60 percent of youth in the labor force have only primary education or less. Even in higher income countries in Latin America and Europe and Central Asia, the majority of youth have secondary education or less (see Figure 3). This not only limits their job opportunities, it also reduces investment prospects in higher productivity and higher value-added activities.4 4 See Almeida et al., (2012). 8 Figure 3: Level of Education among Youth Participating in the Labor Force East Asia and Pacific Latin America and Carribean South Asia Total 0% 50% 100% No education Primary Secondary Post-secondary Source: World Bank Staff, based on International Income Distribution Database Although the focus of policy discussions is often on youth unemployment, this is only part of the problem. Unemployment rates for youth are indeed relatively high in the Middle East and North Africa (9 percent), Latin America (8 percent), and Europe and Central Asia (6 percent); they tend to be considerably higher than those of adult workers (see Figure 4). Far more important is the fact that over 30 percent of youth (20 percent in East Asia and Pacific and Latin America) are idle: not working and not studying (see Figure 5). The share of idle youth has been increasing since the financial crisis. 5 In countries such as Egypt and Lebanon the combined share of unemployed and idle youth has reached 60 and 80 percent respectively. 5 See ILO, IMF, WB report on Jobs. 9 Figure 4: Adult vs. Youth Unemployment Source: World Bank Staff, based on International Income Distribution Database Figure 5: Youth Activity Status by Region East Asia and Pacific Europe and Central Asia Latin America and Carribean Middle East and North Africa South Asia Sub-Saharan Africa 0% 50% 100% Employed Unemployed Inactive in school Inactive out of school Source: World Bank Staff, based on International Income Distribution Database But in the regions with the largest (current and expected) youth populations -- Asia and Sub-Saharan Africa -- unemployment is not the main problem. Part of the reason might be that in lower income settings, youth are less able to afford not to work. The main constraint facing young people in these regions is that they are underemployed, or employed in low- productivity work. 10 In the Middle East and North Africa, South Asia, and Sub-Saharan Africa, between 40 and 70 percent of youth are in the agricultural sector, where productivity is lower and access to social protection schemes unlikely. Among those working outside agriculture, the majority are either self-employed or working without pay (see Figure 6). This does not take into account the fact that, among those who are in wage employment, many are in informal jobs where they also lack access to social security and labor regulations. In Morocco, for instance, 60 percent of youth in wage employment are not registered with the social security. In Latin America, Europe and Central Asia, and East Asia the majority of youth are in wage employment. But again, at least in the case of Latin America and East Asia, employers are often in the informal sector or, if formal, do not offer formal contracts. In rural areas, young people are less likely to have access to land, productive inputs, technology, capital, or markets. They are often consigned to work in occupations that have low productivity and little opportunity for growth and expansion. Rural youth are significantly underemployed, working only a few hours per week. In addition, it can be more difficult for rural youth to gain autonomy: where farms are joint family ventures, young people may be expected to work without pay for the family farm. In some countries, especially where land is held by communal or traditional authorities, young people may not receive equal treatment when access to land is being assigned. In the case of middle income countries, youth are more likely to enter the labor market through the informal sector – particularly those with low levels of education – and take jobs of relatively short duration (see Figure 7). Young people are more likely to leave a job, and they are more vulnerable to being let go from a job, than older workers. This leads to higher overall unemployment rates even if spells of unemployment are not necessarily longer than for adult workers. Eventually, as they growth older and accumulate experience, workers start to move into more stable formal wage jobs or into self-employment. Some research suggests that difficult transitions into stable jobs can have scarring effects that last throughout an individual’s lifetime. There is some evidence indicating that the 11 consequences of unemployment can be more severe for youth than adults. For instance, prolonged unemployment in youth, or delayed entry into work, has potentially large effects on lifetime earnings.6 Prolonged youth unemployment can affect human capital accumulation, physical and psychological health, and careers later in life. For example, youth who take longer to find stable employment likely accumulate less human capital than those who can begin on-the- job learning at an earlier age, and this relative reduction in human capital can lead to lower earnings throughout the life cycle. Long-term unemployment can also erode skills and workforce attachment. Aptitudes and attitudes to perform a given occupation can be lost and human capital and skills can depreciate. Employers may perceive that an individual who spent much of his/her youth in unemployment has some characteristics that make him/her “unemployable,” which can close off opportunities to find jobs later in life. Youth who spend more time unemployed may not acquire the “soft skills” or work habits needed to be effective employees or co-workers, which can lead to fewer, and shorter, employment spells later in life. Unemployment also leads to delays in achieving the productive identity which marks the successful transition from youth to adulthood. Unemployment experienced at early ages is associated with discouragement, illness, stress, and depression in later life as well as with lower life expectancy.7 Difficulty in entering the labor market can lead to feelings of hopelessness and low self-esteem, especially among those who are in the process of forming occupational identities.8 Unfortunately, most of this evidence comes from developed countries. There are almost no long-term studies of prolonged unemployment in middle- and low-income countries. There is also little evidence concerning the extent to which young people are scarred by early labor market setbacks, or what the corresponding social costs might be. 6 See Kahn (2010), Gregg and Tominey (2005). 7 Bell and Blanchflower (2010). 8 Bowman (1990). 12 Figure 6: Youth employment Status across Regions Source: World Bank Staff, based on International Income Distribution Database Figure 7: Labor Transitions among Youth in Brazil and Mexico Brazil Mexico Source: World Bank Staff based on Pesquisa Mensal de emprego (PME) and Encuesta Nacional de Ocupación y Empleo (ENOE) 13 These individual-level effects can have significant consequences for firm-level and national productivity and economic growth. Significant youth unemployment represents immobilized labor resources that are not contributing to national output. Moreover, when youth acquire less human capital due to extended periods of unemployment, it affects productivity for the economy as a whole that can restrain growth below its potential. Indirectly, when employers are faced with a young workforce that lacks important skills, the set of profitable production technologies shrinks and investment decisions are affected, thus potentially setting the country on a slower long-term growth path. A difficult school-to-work transition for youth may also have broader implications for social cohesion and violence. It has been claimed that unemployed youth provide fertile grounds for gangs, violent resistance movements, and criminal activities. Without regular employment to provide structure to their daily lives, some youth can slide into activities that are disruptive to society.9 They also may demonstrate less civic engagement and disrupt otherwise well-functioning collective governance. But there is still limited knowledge about these issues and no evidence yet that unemployment among young people necessarily leads to violence.10 3. A framework for action on youth employment Ultimately, labor market outcomes of interest – unemployment, inactivity, employment status – depend on individual decisions to participate in the labor force by engaging in wage employment or self-employment/entrepreneurship, as well as firms’ decisions about whether to create new or expand current establishments or business. These decisions and their outcomes are affected by individual characteristics (e.g., skills, social networks, psychology); the functioning of product, labor, and capital markets; and macroeconomic, demographic, and geopolitical factors that affect the level of aggregate demand and the total supply of labor. 9 See e.g. Cincotta et al. (2003). 10 Cramer (2011). 14 All of these factors are clearly relevant for adult and young workers alike. It is possible, however, that young workers are more vulnerable or responsive to all of them, which could then justify targeted interventions. Adult workers with a low level of education, for instance, can make up for initial skills deficits through work experience. When regulations increase the minimum cost of labor or the cost of dismissing workers, employers might prefer not to take risks with inexperienced young workers and hire adults instead. Young workers are also less likely to have access to capital to initiate a business and can be more affected by contractions in aggregate demand (see Table 1).11 This section has two objectives: first, to start building a map of the links between individual, market, and macro factors and youth labor market outcomes, based on available literature; and second, to identify the types of interventions that can influence these three sets of factors and/or their effects on labor market outcomes as a prelude to the analysis presented in the next section. In doing so, the section also discusses, from a normative perspective, when interventions targeted to youth vs. all workers make sense and the types of information that would be needed to inform the decision. 11 See Cho and Newhouse (2010). 15 Table 1: Constraints to Youth Employment Type of constraint Youth Only or All Workers Inadequate cognitive, non-cognitive and technical skills Ya Individual Lack of social networks Ya Household income A Psychological biases Ya Distorting labor regulations Ya Markets and Government Low bargaining power Ya Failures Lack of information about where the jobs are Ya Employers lack information about workers Ya Lack of access to capital Ya Constraints to business creation and job creation Ya Economic slowdown Ya Macro Age cohort-size Y Conflict and violence Ya Notes: Y=constraint affects only youth; Ya=affects all workers, but youth more than others; A=affects all workers. 3.1 Individual Characteristics Skills--technical, cognitive, and non-cognitive--are probably the most important factors determining labor market outcomes. Skills first determine decisions about labor market participation. Other things being equal, skilled workers, including women, are more likely to participate in the labor market, in part because the opportunity cost of not working is 16 higher. In Lebanon, for instance, women with a higher education diploma are 20 percent more likely to enter the labor market.12 Skills also determine the choices that individuals have in the labor market, either as self- employed workers, entrepreneurs, or wage employees. High productivity/high earnings jobs in the formal sector are often reserved for the graduates of the best universities and/or workers with very specific technical and professional skills, usually acquired in graduate programs. Opportunities for those without college degrees or only secondary education are more limited, hence, the strong correlation observed across countries among the level of education, occupation, and earnings. But having a diploma is not a guarantee of success. In India, for instance, the share of youth with at least a secondary education will more than triple over the next two decades. Yet today, 50 percent of university graduates obtain a diploma in arts, far exceeding employer demands.13 In Tunisia14 and Lebanon,15 more than 50% of university graduates are in jobs that do not use the skills they acquired in university. In Cambodia, employers also complain about skills mismatches, especially among out-of- school youth.16 Moreover, employers are not necessarily concerned that young workers have inadequate formal education.17 Rather, there is growing evidence of the importance of non-cognitive or behavioral skills in determining labor market outcomes, even after controlling for their effect in the acquisition of other skills. Surveys of employers often find that they are looking for individual-specific characteristics such as honesty and the ability to solve problems, communicate, and work in teams.18 The right set of non-cognitive skills is also essential during the job-search process or when starting a business. Writing and communication skills, perseverance, timeliness, and personal presentation skills are, for instance, critical when preparing job applications, 12 See World Bank (2012). 13 See Robalino et. al. (2011). 14 Skills toward Employability and Productivity (StEP) framework in Banerji et al.(2010). 15 See Robalino et al., (2012). 16 See Almeida et al., (2012). 17 See www.enterprisesurveys.org. 18 See Heckman et. al. (2006), Blom and Hobbs (2007). 17 conducting interviews, or negotiating salaries. Certain cognitive skills and preferences (e.g., risk-taking) are preconditions to becoming a successful entrepreneur, and these are likely to be acquired early in life. In fact, many youth who enter self-employment might do so not because they have the aptitude, but rather because there are no other options. 19 Household income can be an important determinant of participation rates, occupational choices, and employment status. For many youth, a premature entry into the labor market can result from financial constraints related to the direct cost of education (including transportation when there are no schools nearby) or the opportunity cost of not working. The ability to start a business or the flexibility and time youth have to find a job are also highly correlated with the income of their parents or close relatives. Grants or credit from family members can help overcome credit constraints in capital markets (see next section), and when rent, food, and transportation expenditures are covered youth can wait to find the job they really want and reject offers they consider inferior. 20 This can, of course, increase unemployment rates – unemployment becomes voluntary – but at the same time lead to better matches. Social networks, like skills, can affect the labor market opportunities that youth have, but also influence the beliefs and attitudes towards work and different types of occupations. Job opportunities, including self-employment opportunities, depend on the cues, information, and advice that young people receive from family and friends. Youth in low- income families can be less well-connected than youth in high-income families. In some countries, social networks can also influence the likelihood of entering self-employment or becoming an entrepreneur, as an absence of contacts can make the creation of certain businesses difficult. In other countries, the reverse may be true: the only role models around are self-employed, and that may be the only option that young people see for themselves. In such a context everyone, regardless of capacity or aptitude, wants to be an 19 See Margolis et al., (forthcoming). 20 See World Bank (2012). 18 owner-operator. Youth can also be influenced by the choices of other peers and engage in activities that have limited economic prospects or that are illegal.21 Psychological factors are, unsurprisingly, key determinants of the decisions youth make at all levels, and these decisions are not always in the best interests of the young people who make them. Decisions about investments in education and career choices are complex and involve comparing multiple choices with uncertain outcomes.22 The heuristics that youth use to solve these problems can be affected by lack of information, biased expectations about the future, high discount rates, or time-inconsistent preferences.23 When students choose a specialization for technical or vocational education or a major in tertiary education, they may lack, or fail to process, the information about the implications of their choices for their future careers. Youth may also have biased expectations about the implications of stopping their education too soon and their opportunities in the labor market. Procrastination, lack of self-control, and lack of self-esteem can also influence the behaviors of young adults leading to underinvestment in skills, low effort and too little time invested in job-search activities, or simply withdrawal from the labor force and idleness in the face of rejection.24 3.2 Market and Government Failures The likelihood of obtaining a good salaried job or being able to start a business is affected by how well labor and other markets work, which in turn depends, in part, on the types of labor and business regulations that governments implement. Common problems in labor and capital markets are linked to limited information that can affect the hiring process or restrict access to capital, thus affecting transitions into self-employment. Barriers to entry in product markets can also lead to situations where employers have more bargaining power than workers, particularly youth, and can therefore impose inappropriate working 21 See Akerloff (1997). 22 See Tahler and Sustain (2007). 23 See Ariely (2010). 24 See Almeida et al. (2012). 19 conditions or depress wages. A majority of countries have implemented regulations to address these problems, but sometimes these can produce more difficulties for youth. Information and labor markets. As discussed above, youth may lack adequate information about job opportunities, and employers may not have enough information about the skills and behaviors of young applicants, particularly when they are first-time job seekers. Applicants may hold diplomas, but these do not necessarily credibly certify or signal to employers the technical skills that they have. In terms of behaviors, employers sometimes expect that youth are less reliable, have less-developed communication skills, and are more mobile and therefore less likely to stay in their jobs. Hence, hiring young workers with no previous work experience, and therefore no professional references, can be a risky bet. Other things being equal, employers will prefer to hire adult workers. Information and capital markets. Youth may also lack the financial capital needed to enter self-employment and/or entrepreneurship, as they have had less time to accumulate savings or assets than older workers. Youth are also less likely to be living independently, with fewer physical assets in their name that can be pledged as collateral. Finally, banks may consider youth to be at high risk of non-repayment and charge higher rates for young people to borrow even when assets are available to pledge as collateral. Regulations. Employment protection legislation, while designed to protect workers in existing jobs, can have negative effects on the likelihood of youth finding new jobs. Situations may arise where employers desire skills sets that youth may possess but their incumbent workforce does not (for example, cutting-edge programming skills). In this case, employers might find it desirable to replace some existing workers with youth that have the desired skills, but employment protection legislation may make separating from existing employees too costly, or even impossible, thereby reducing the chances of skilled youth finding jobs. There is evidence that, in general, regulations on hiring and dismissal procedures that are too rigid can discourage hiring, as employers prefer not to have to incur the costs of 20 dismissing bad matches. These regulations can have a higher effect on young first-time job seekers who, as discussed above, are a riskier bet for the employer.25 Regulations such as minimum wages and mandatory benefits financed through payroll taxes can also push youth out of the formal labor market. Indeed, employers can hire workers as long as their marginal contribution to profits is positive: the extra earnings from additional labor are higher than the additional expenditures on wages. If regulations increase the minimum cost of labor, workers whose productivity is not high enough to generate the extra earnings needed to justify their salary may suffer – unless they are hired “informally” with a salary below the minimum and/or no enrollment into the social security. This is more likely to be the case among unskilled workers and youth, since productivity is usually correlated with age. This seems to be the case, for instance, in Tunisia which has a relatively high minimum wage and tax-wedge.26 High wages and benefits in the public sector, or generous social assistance or unemployment benefits, can also reduce incentives to search for or take jobs. This is evident, for instance, in the cases of Egypt and Tunisia in which the public sector offers high salaries and good benefits.27 These advantages can provide incentives for youth to queue for jobs in the civil service and lead to an increase in unemployment rates. There is also evidence that generous unemployment insurance can prolong unemployment spells, at least in the case of high income countries.28 Although very few youth have access to unemployment benefits in the case of middle- and low-income countries, there have been several initiatives, particularly since the financial crisis, to provide grants to individuals who are participating in training programs or searching for jobs. This was, for instance, the case of the Tunisian “Hope” (AMAL) program. However, if not well-designed and if the conditionalities are difficult to enforce, this type of intervention can reduce incentives to search for jobs and accept job offers. 25 Montenegro and Pages (2003). 26 See Rutkowski (2012) and Robalino et al. (forthcoming). 27 See MENA report on Jobs. 28 See Fiesben and Chadi (2008). 21 Other markets. Young people may also be constrained by the poor performance of markets for land, inputs, goods, and services. Especially in rural Asia and Africa, distances between producers and consumers, the time between input and harvest, and the perishability of commodities mean that agricultural markets are prone to high transactions costs and significant risks. These risks may be exaggerated for younger farmers, who in any event express a strong preference to migrate to urban areas, citing lack of access to land, credit, and inputs, as well as poor social amenities in rural areas, and poor returns to farming.29 3.3 Macroeconomic, Demographic, and Geopolitical Factors Labor market outcomes for youth and other workers are, to a large extent, determined by the entry of new businesses and firms’ decisions to create new or expand current establishments. Corruption, excessive bureaucracy and red tape, underdeveloped capital markets, lack of infrastructure, and other features of the business environment can hamper the process of business creation (and destruction), leading to low job-creation rates and fewer vacancies relative to the number of job-seekers – an increase in the “tightness” of the labor market. Finding a job therefore becomes more difficult for the average job-seeker. But given the constraints discussed above, labor market tightness is likely to disproportionately affect young workers. In the United States, for instance, even after the end of the last recession, job creation rates have remained low – following a trend that started in the early 90’s (see Figure 8). Unemployment rates have fallen only slowly, particularly for youth. 29 See e.g. Akpan (2010). 22 Figure 8: Job Creation and Destruction Rates in the United States Source: Haltiwanger (2011) Downturns resulting from insufficient credit or labor demand can also disproportionately affect youth. These downturns affect labor markets both by reducing job creation rates and increasing job destruction rates. During the last financial crisis, among active workers, youth experienced the largest adverse impacts on employment, unemployment, and wage employment.30 Part of the explanation is that, as labor markets tighten, it becomes more difficult for youth to find jobs. But firms’ layoff decisions can also make a difference. In particular, firms may find young and unskilled workers more dispensable because they have acquired fewer important firm-specific skills.31 Alternatively, employers might prefer to dismiss young workers first when wages cannot be adjusted downwards because of minimum wages. Demographics can also play a role in determining labor market outcomes for youth. Historically, many countries have gone through periods when, as a youth bulge enters the labor market, unemployment rates among young workers increase.32 Even if there were no differences between young and adult workers, and therefore they had the same 30 See Cho and Newhouse (2010). 31 World Bank, 2009 [cited in Cho and Newhouse (2010)]. 32 See Layard et al., (2005). 23 unemployment rate, an increase in the relative size of an age cohort can drive the unemployment rate up and wages down.33 In essence, if there are more youth entering unemployment for the first time as they search for their first job, even if exit rates are the same as for adults, unemployment rates for youth will become higher. 34 Demographic factors may thus add to the challenges facing youth in the labor market in countries where the number of young people is rapidly growing, such as Sub-Saharan Africa, South Asia, and the Middle East. Lastly, situations of conflict and violence tend to disproportionately affect youth and thus their perspectives for employment. Security and safety concerns may be present in conflict environments, and parents may attempt to shelter their children from violence by limiting their mobility (and thus ability to find and occupy a job). Moreover, in conflict situations youth are often conscripted to fight. Transiting into productive employment can then be difficult for the same reasons that affect the long-term unemployed; while conscripted, youth fail to acquire or even lose the necessary technical, cognitive, and non-cognitive skills for productive employment. 3.4 Policies and Programs Programs and policy responses have been developed to address some of the constraints that adult and young workers face to access jobs. Table 2 briefly summarizes the main initiatives, some of which are reviewed in the next section. Training programs, for instance, can be used to upgrade life and technical skills. Employment services can be used to overcome information problems in the labor market. Skills certification programs can address problems related to signaling. Wage subsidies can allow youth to acquire some work experience and improve their employability. Regulatory reform can also address some constraints to employment by better aligning the costs and benefits of employing youth. 33 Ibid. 34 The net effect on employment will be a function of the response by employers to lower wages among higher-productivity youth. Lower wages may spur investments in technology to make use of the labor pool, as among the East Asian “Tigers” [see Bloom and Canning (2005)). 24 An important question, however, is whether governments should have specific policies to help youth as opposed to workers in general. At least in theory, the implementation of any of the programs listed in Table 2 would be justified if the increase in social welfare resulting from better labor outcomes is higher than the social cost of implementing the programs (and if there are no other interventions that can generate higher increases in social welfare for the same amount of resources). In non-targeted programs, one can assume that youth and adult workers benefit in proportion to their share in the labor force or the share of the total number of unemployed. A program targeted to youth would not only exclude adults, it may also worsen labor market outcomes for them (e.g., wage subsidies for young workers can penalize adult workers by inducing substitution). In principle, targeting can be justified if the net increase in social welfare resulting from improving labor market outcomes among a group of youth is higher than what could be obtained by focusing on a group of the same size but which includes adult workers. One way this could happen is if the present value of benefits related to improved labor market outcomes among youth is higher than among adults. For instance, a program could reduce unemployment more among young workers or the increase in productivity could be higher. Another justification for targeting would be that, as discussed above, the social costs of poor labor market outcomes among youth are higher than among adult workers. This is, of course, difficult to assess. What should countries do? Much will depend on the context, social preferences, and priorities of each country. In countries with very high youth unemployment rates, such as Tunisia and South Africa, not having targeted programs would be politically unsustainable. In general terms, however, it would seem advisable to operate on two fronts. First, countries should develop policies and programs that aim to improve labor market outcomes for all workers, including those policies that improve the business environment and facilitate the process of job creation and destruction. Active labor market programs such as employment services and training programs would be open to all workers in this setting. In 25 parallel, there could be special interventions that target vulnerable youth, including school dropouts and those living in low income and poor households. Table 2: Policy and Program Responses to Constraints to Youth Employment Type of Constraint Policy or Program Response Inadequate cognitive, non-cognitive and Training, Education reform technical skills Lack of social networks Mentoring, Intermediation Household income Microfinance, Loan guarantees, Social Safety Micro Nets Psychological biases Public sector reform, Social assistance reform, Incentives to foster development of a private domestic services sector, Publicity Campaigns, Early-Childhood interventions, Anti- Discrimination Laws Distorting labor regulations Labor code reform, Regulatory reform, Standards reform Low bargaining power Intervention in collective bargaining, Minimum wages Market Lack information about where the jobs are Intermediation, In-School Counseling, Job Fairs, Help Wanted Web Sites Employers lack information about workers Intermediation, National Qualifications Frameworks Lack of access to capital Microfinance, Loan guarantees Constraints to business creation and job creation Business Environment Reform, Wage subsidies, Infrastructure investment Slow economic growth Expansionary macroeconomic policy, Trade Macro policy Age cohort-size Investment in educational capacity, (Targeted) sectoral or regional growth incentives Conflict and violence Improved Security 26 4. What have countries been doing? What has worked? Governments and NGOs have addressed the constraints discussed above using a variety of programs35 that have been catalogued in the Youth Employment Inventory (YEI).36 The YEI is a global knowledge repository that currently includes information on 500 youth employment programs from around 100 countries, documenting program design, implementation, and achieved results. Programs covered by the YEI database range from interventions for improving the labor market information base, to programs that aim at increasing the demand for youth labor (wage subsidies and public works), to those focusing on improving chances for young entrepreneurs (providing financial, technical, and training assistance). They have been classified into five categories: (i) skills training (including vocational training, on-the-job-training programs, literacy and numeracy programs, second- chance and equivalency programs, and soft-skills programs); (ii) entrepreneurship promotion (financial assistance, technical assistance, and entrepreneurship training); (iii) subsidized employment (including wage subsidy programs, public works, and public/community service programs); (iv) employment services (including search assistance and access to labor market information, job counseling and placement services, and financial assistance for job search); and (v) reforms to labor market regulation (including anti-discrimination legislation). Although the programs are used to varying degrees in different regions, training programs are by far the most common interventions in the YEI database for all regions.37 Training 35 A good resource for understanding active labor market programs for youth is Cunningham et. al. 2010. 36 The Youth Employment Inventory (YEI) is a joint effort of the World Bank’s Human Development Network/Labor Team (WB/HDN), the German Ministry of Economic Cooperation and Development (BMZ), the Inter-American Development Bank (IADB), the International Labour Organization (ILO), and the Youth Employment Network (YEN). 37 Inclusion of programs in the YEI database did not follow any specific selection criterion; rather, partners had a tendency to disproportionately enter programs for which they were responsible. As the portfolios of the ILO and BMZ lean more heavily toward training programs than, for example, labor market reforms, the database should not be considered a random sample of available programs and some over-weighting of training programs among interventions is to be expected. 27 represents 86 percent of the 500 cases currently included in the database (see Figure 9).38 At the other extreme, labor market reforms are the least frequent intervention, corresponding to only four percent of all cataloged interventions. Entrepreneurship promotion worldwide is used in 39 percent of the cases, followed by employment services (34%), and subsidized employment (12%). Figure 9: Youth Employment Policies and Programs, by Region and Type of Intervention Source: YEI. There is some variation across regions in the types of programs implemented. Among programs in the database, employment services are relatively more popular in the OECD, entrepreneurship promotion in Sub-Saharan Africa, LAC, and MENA, and subsidized employment in ECA and the OECD. This variation in program responses could be due to different sets of constraints to youth employment in different regions. Alternatively, the variation could be due to differences in the amount of information countries have about what programs should be implemented and how to implement them. Other variations could arise from different capacities to implement (for example, employment services tend to be more frequent in middle-income countries (MICs) and high-income countries (HICs) than in 38 Although focused on a narrower time frame and using a different selection criterion, the joint ILO/World Bank Inventory of Policy responses to the Financial Crisis (http://www.ilo.org/crisis-inventory) finds that “about 78 percent of reported policy measures focused on the supply side,” essentially through training [Mahmood et. al. (2012)). 28 low-income countries (LICs), partly due to the difficulty of maintaining a network of highly functional employment offices); political economy considerations (labor market reforms may be harder to implement in thriving democracies where the losing parties have a voice and policymakers fear for their jobs); or budget capacity (public works can be relatively expensive).39 A final possibility is that the variations are due to selection criteria in the way programs are entered into the database, with partners who operate more intensely in certain regions with specific types of programs being more likely to include those programs in the YEI. 4.1 Training Programs Training programs mainly address the market failures that result in an inadequate supply of skills – technical, cognitive, and non-cognitive.40 While vocational training programs and technical skills interventions specific to certain sectors have been used by policymakers for many years, more programs today combine cognitive skills training with non-cognitive/soft- skills training. The curricula of these programs include subjects such as teamwork, leadership, communication skills, positive attitudes, confidence in self and others, respect for self and others, problem solving, work organization, and knowledge related to risky behavior (violence, reproductive health, drugs, etc.). Communication skills, planning and organization of work, customer service, health and safety at work, environmental protection, and guidance for job search have also recently entered into some programs’ curricula. Moreover, comprehensive interventions that group training with other services (employment services, subsidized employment) have become increasingly popular in recent years. 39 Note that this is in terms of budgetary cost, and not opportunity cost. It is important to express policy alternatives in terms of opportunity cost as well as budgetary cost: an “expensive” program may save money later, or it may be more cost-effective than alternative programs that appear cheaper from a budgetary perspective. 40 Heckman et al. (2006) show both cognitive skills and non-cognitive skills are important in explaining labor market outcomes. 29 The target populations for training interventions are not always restricted to the unskilled and low-skilled. While around 90% of youth programs in OECD countries target the low- skilled, it is close to 60% in Latin America, while for Africa, Eastern Europe, and Asia, only 30% seem to target the low-skilled youths.41 In MENA, government training programs often include all young people (often encompassing age brackets beyond 27 years of age) without a specific focus on the low-skilled and/or focus specifically on youth transitioning from school to work (including those with tertiary education).42 A typical example is the AMAL program in Tunisia, which aims to promote employment for long-term unemployed tertiary education graduates who have yet to transition to stable employment by means of a training-plus intervention. The program was designed to offer participants a stipend equivalent to 80 percent of the minimum wage and a series of services, ranging from training in life skills and technical skills, counseling, job-search assistance, and wage subsidies. In practice, 20,000 participants received the stipend along with various services, while an additional 120,000 received the stipend alone. The effectiveness of youth training programs tends to be higher when classroom training is combined with internships in firms. On-the-job training is an important channel through which first time job-seekers receive and upgrade relevant skills to become competitive in the labor market. Studies compiled in the YEI show that on-the-job training yields large and positive impacts on the trainee’s earnings and productivity both in developed and developing countries.43 Furthermore, total hours spent in the classroom training have a lower impact on the program’s rate of return than the time spent with on-the-job training. The intensity of the training provided is also an important determinant of success. Comprehensive training programs targeting unskilled youths have been applied in Argentina (Proyecto Joven), Chile (Chile Joven), Colombia (Jóvenes en Acción), the Dominican Republic 41 Fares and Puerto (2009) and preliminary estimates from the YEI database. 42 See for instance Angel-Urdinola, Semlali, and Brodmann (2010). 43 See Almeida and Cho (forthcoming) for a comprehensive review of the effects in developing and developed countries. 30 (Juventud y Empleo), Peru (Projoven), and Panama (PROCAJOVEN).44 These programs typically provide for many months of intensive soft-skills and technical classroom training, employment services, and on-the-job training coupled with subsidized employment/internships, and they tend to be more successful than versions of training programs with a similar structure but fewer weeks/less comprehensive training and more generous employment subsidies, like in Jordan’s NOW program.45 Interesting new comprehensive training programs encompassing socio-economic behavioral objectives are currently funded by the Adolescent Girls initiative. An example is the Economic Empowerment of Adolescent Girls (EPAG) project in Liberia which provides demand-driven technical skills coupled with business development and life skills for out-of- school girls aged 16-27, with the goal of increasing the economic well-being and socioeconomic behavior of young women. The training is further linked to performance- based incentives for job placement services and participants are encouraged to form informal rotational savings clubs and invest in micro-enterprises. Preliminary results from the mid-line survey suggest that the program led to a 50 percent increase in employment and a doubling of incomes among the participating young women, compared to a control group. Interestingly, the increase in average income resulted from bringing more young women into employment (on the extensive margin) rather than increasing average incomes among those who were already working (the intensive margin).46 Common characteristics of successful training programs also include responsiveness to market signals and attention to incentives and governance considerations for training providers and participants. Many innovative programs outsource the provision of the training services to private training institutions/NGOs through a competitive bidding process, especially when national employment service systems have low capacity or are 44 See Ibarrarán and Rosas Shady (2008) for evidence on Jovenes programs. See Rodrigues-Planas and Jacob (2008) for an overview of the evidence in transition countries. 45 Groh et. al. (2012). 46 Note that greater success was seen among those who received self-employment training than among those who received job skills training, although both were significantly higher than the control group [World Bank (2012b)). 31 ineffective. While competition between private sector firms/training providers and public sector training institutions has been established over the past decade(s) in most OECD countries, this is a more recent phenomenon in developing countries. In Lebanon, for instance, the New Entrants to Work Program (NEW) selects several competing NGOs/private sector firms on a competitive basis to provide (life-) skills training and participant placement into positions that offer on-the-job training (which is combined with a wage subsidy). In addition, a bonus/reward system based on successful completion of the classroom training and/or the on-the-job training (OJT), monitored through a specifically designed IT-registration management system, serves as a vehicle to ensure the quality of the different components. Training programs for youth can affect outcomes other than employment and income. For example, India’s Livelihoods Training for Adolescents Living in Slums of Allahabad program, targeted at adolescent girls aged 14-19 in the urban slums, provides participants with vocational training and counseling, while also informing them about reproductive health. In addition, beneficiaries received assistance in opening savings accounts. Although the girls in the treatment group were more likely to have better social skills, be part of a group, and have greater probability of being informed about safe spaces and reproductive health than those in the control group, the program failed to have any significant impact on labor market outcomes, self-esteem, mobility, and gender-role attitudes. 4.2 Entrepreneurship Promotion Programs Entrepreneurship promotion programs can combine several types of interventions to support successful self-employment. The programs may include interventions such as training, providing access to finance, advisory services and mentoring, microfranchising, enabling value chain inclusion, small business networks, support for technology transfer, business incubation, and many others. Curricula of the training programs cover a variety of subjects from business skills tailored to self-employment activities and financial literacy to entrepreneurial and life skills. The access to finance programs often involve micro-finance 32 organizations and lending programs (like FINCA, BRAC, or ACCION International) and can have added modules of training on business skills or financial literacy. The advisory services provided by these programs can include mentoring, counseling, or assistance with business formalities. Microfranchising, another promising – yet unevaluated - program, typically provides beneficiaries with comprehensive support to enable them to replicate an existing and proven business model, often with the help of training, access to finance, networking, mentoring, and many other components. Lastly, value chain inclusion (sometimes referred to as value chain integration) aims to establish sustainable linkages between existing producers and other actors in the value chain. The target groups for entrepreneurship promotion programs are almost as diverse as the programs themselves. Entrepreneurship promotion programs such as Junior Achievement may target in-school students; other programs (ILO’s Start and Improve Your Business or the programs under Youth Business International) target potential entrepreneurs who are out of school; and still other programs target existing entrepreneurs (Business Edge, IFC). Beneficiaries may be living in rural (targeted by TechnoServe) or urban areas (targeted for instance through Infrastructure Resource Center IRCIRC). Some of them have a special focus on women (India’s Self Employed Women’s Association SEWA), or other vulnerable groups such as street children (Street Kids International). In fact, while many programs focus on out-of-school populations, numerous countries, especially in the OECD, offer entrepreneurship education and training early in the formal schooling system. Moreover, there is an ongoing debate over whether entrepreneurship promotion programs should target high-growth-potential entrepreneurs (known as “gazelles”), which account for a marginal number of self-employed but whose firms have the potential to provide significant employment to others, or the self-employed and small-scale entrepreneurs, who make up the vast majority of entrepreneurs yet are less likely to grow.47 47 Michael Grimm, Peter Knorringa, Jann Lay (2012), “Constrained Gazelles: High Potentials in West Africa’s Informal Economy,” World Development, Volume 40, Issue 7, July 2012. 33 In rural areas, attention has recently shifted to the creation and exploitation of agricultural value chains, in which young entrepreneurs are given training and the inputs required to engage in high-value-added activities, such as post-harvest processing and distribution of agricultural goods. More importantly, it encourages young entrepreneurs to consider the entire chain of supply and demand in the production and distribution of their products, and to engage in activities in value chains that are competitive and growing. For example, a local development project in Vietnam worked to improve the returns to young rattan producers by improving technical skills and productivity as well as quality, branding, and enhancing linkages to national and global markets, including firms that supply IKEA.48 In sub-Saharan Africa, where more than half of all young people live in rural areas, programs such as the Junior Farmer Field and Life Schools (JFFLS) of the FAO and ILO’s Start and Improve Your Business (SIYB) provide technical training, business skills, and life skills, and assist young people to negotiate credit, land, and product markets.49 Although the evidence is still scarce, preliminary results of a meta-analysis of impact evaluations50 of entrepreneurship promotion programs show that overall, “comprehensive” programs perform better. In particular, entrepreneurship training complemented with financial support improves labor market activity significantly more than either of those two components delivered separately. The Youth Opportunities Program (YOP) component of the Northern Uganda Social Action Fund (NUSAF) is an example of one such comprehensive program that has had significantly positive results.51 Targeting the most vulnerable unskilled youth (aged 15-35 years old) who are unemployed or under-employed in 29 of the most vulnerable districts of Northern Uganda, YOP required youth to organize themselves into groups and identify a common vocational skill of interest. The program then provided vocational training relevant to starting a new business, life skills training, and psychosocial counseling, accompanied by the opportunity to receive a cash transfer to finance this 48 International Labor Organization (2011). 49 The SIYB has reached millions of people since 2003. See Van Lieshout et al. (2012). 50 Cho and Honorati (2012). 51 Bigirimana and Namara (2010) and Muzi et al. (2011). 34 training as well as advisory services. As a result, the average beneficiary was nearly 100 percent more likely to be engaged in skilled employment, and saw hours spent on market activities increase by roughly a third, relative to non-beneficiaries. Profits increased by nearly 50 percent on average. Modest improvements in social cohesion and community participation were also observed. Entrepreneurship promotion programs have also been found to be effective for improving intermediate outcomes. Impacts on business knowledge and practices are more likely to be positive and significant than on final outcomes such as earnings and profits. Attitudes and traits such as risk aversion, confidence, willingness to compete, optimism, and time preferences can also be affected by entrepreneurship promotion programs, as can savings and borrowing behavior. The effectiveness of programs is influenced by the context in which they are implemented. Programs implemented in countries with a large agriculture sector and low education levels appear to be more successful. Stand-alone training programs are more likely to have larger impacts in countries that are urbanizing. Financing works better in low-income and lower- middle-income countries than upper-middle-income countries, and when the financial costs of starting a business are high, interventions without financial support are not as effective. 4.3 Employment Services The term “employment services” covers a range of intermediation -related activities designed to address market-level employment constraints. The least-intensive employment services involve collecting and reposting job offers or individual CVs. More sophisticated services include job search assistance, CV writing and interview assistance, and direct intermediation/placement services. Some countries also provide in-depth coaching, counseling, and orientation services. Each level of service provision requires increasing capacity on behalf of the service provider, and in some cases service provision is outsourced to the private sector. More recently, technology-based information sharing systems have 35 shown to be a promising, cost-effective way to outreach to youths and clients in areas with low capacity/infrastructure. In order to be effective, the types of employment services offered should be aligned with the capacity to provide them at a sufficiently high level of quality. In low-capacity countries, the databases of job postings may not be maintained and the information can easily become out-of-date and worthless. Attempts to provide sophisticated services require significant resources; a lack of sufficient resources was at the origin of the failure of Tunisia’s AMAL program to be implemented as planned. On the other hand, employment offices tend to be full service centers in countries with high capacity. For example, the United Kingdom’s Job Centers Plus offer a range of services, including career counseling, training or education program placement assistance, job matching, labor exchanges, and other related services. The centers also merge labor market advisory with social services, integrating benefits and job search in order to better reach people furthest from the labor market. In Bulgaria and Azerbaijan, local employment offices also operate on the “one-stop shop” model, with front offices that handle routine initial queries (and requests to certify unemployment – needed for social assistance applications), and “back offices” for higher -end services for jobseekers and employers.52 Clients are profiled for skills, education, qualification, and needs by a trained interviewer who immediately enters the data into the MIS database of jobseekers. The local office and the job seeker also jointly fill out a tailored “mutual obligations” personal plan to signal co-responsibility for finding a job. The clients then access a range of activation services, including: (a) training for various skill levels; (b) counseling and job- readiness advice (for the hard-to-serve); and (c) intermediation and job brokerage services. The centers interface not only with jobseekers but also with employers, who are offered services such as hosted recruitment events. Azerbaijan is experimenting with a similar model in its Sumqayit local employment office.53 52 Kuddo (2009). 53 Kuddo (2009). 36 Technology-based information sharing is a promising new way to foster cost-effectiveness of intermediation services and broaden the outreach to youths and clients in low-income settings with low capacity and infrastructure. Instead of physical job offices, use of internet, radio, or cell phone text messaging may provide an opportunity to offer employment services across larger areas. For example, the Palestinian-Canadian NGO and mobile service SoukTel has developed a SMS-based “job matching” application that is currently being applied in the Palestine territory, in Morocco, and in East Africa. The system provides a platform for firms and potential employees to connect via text messages in places where young people are far more likely to have a cell phone (approximately 85% in most developing countries) than internet access (approximately 35%). There is no evidence to date involving a control group. However, SoukTel’s growth experience and financial self - sustainability make it an attractive option; it is location-independent, empowering youths in rural areas lacking infrastructure, and time-independent, without the need for costly internet or long distance calling.54 Programs to support the job-search and matching process should not discriminate based on the legal status of the firm. Excluding the informal sector will drastically reduce the number of jobs advertised. For example, private training providers in the Life Skills Education for Employment and Entrepreneurship program in Indonesia train, certify, and match unemployed youth with domestic and overseas jobs in either the formal or informal sectors.55 When implemented well in a high-capacity environment, employment services tend to be relatively inexpensive and cost effective, particularly for the short-term unemployed. In Romania, employment services were found to be among the most cost-effective interventions implemented in the late 1990s from within a broad set of active labor market 54 Weber (2012, forthcoming). 55 Table 3 in this document reflects existing intermediation services and does not include the new technology- based information sharing via cell phones and similar applications. 37 policies, followed by training and by wage subsidies.56 Peru’s “Proempleo” is another example of a well-run and diverse program reaching vulnerable low-income youth, low- skilled workers, and the disabled. The cost of the program is estimated at only 0.4% of GDP.57 Evidence from the New Deal for Young People in the United Kingdom, which provided a combination of wage subsidies to the private sector and job search assistance, shows that approximately one fifth of the effect was explained through improved job search assistance (instead of the wage subsidy component).58 4.4 Subsidized Employment Programs Subsidized employment programs come in two main types: wage subsidies and public works. Unlike the situation with adults, where wage subsidies can be used to protect existing employment,59 subsidized employment programs for youth typically focus on moving young people into jobs that they did not already hold. Wage subsidy programs can operate in many ways, either through a direct transfer to employers, reductions in payroll taxes, or income tax credits. These programs aim at promoting participation in the labor market of the beneficiaries, while compensating employers for screening, orientation, and initial training costs. Wage subsidies may also be given to employers to stimulate the demand for specific workers or to employees to provide incentives for re-employment. Public works programs, on the other hand, provide the job and (most often) compensation directly to the individual. These programs were an important tool in mitigating the impacts of the recent 56 In Romania, employment services provide job and social counseling, job search assistance, job placement services, and relocation assistance. In the late 1990s and early 2000s, beneficiaries had access to the services up to nine months and were entitled to up to two months of salary at the ongoing minimum wage. Implementation of ALMPs is done by the National Agency for Employment and Vocational Training and the county agencies for Employment and Vocational Training. 57 Proempleo provides advisory services to jobseekers as well as employers. Jobseekers receive information about job openings but can also request advisory services (counseling) for career choices and support in preparing Curriculum Vitaes (CV) and preparing for interviews. If appropriate, jobseekers can also be referred to alternative training programs to improve their employability. Employers get access to a database of jobseekers and receive services, including the prescreening of potential candidates and support with official administrative procedures related to preparing and registering labor contracts. 58 Blundell et al. (2004). 59 An example is the Kurzarbeit program in Germany during the crisis; see Mahmood (2012). 38 financial crisis, and they can be managed directly by the government or contracted out to NGOs or the private sector. The target populations for subsidized employment programs tend to be more vulnerable youth and those who are further away from being able to obtain employment without assistance. For example, Lebanon’s New Entrants to Work (NEW) program offers vouchers to unemployed first-time job seekers that cover the cost of social security contributions during an on-the-job training period, with the idea that young workers without prior job experience are perceived as less productive than those with work experience, and will thus need the additional assistance to find a job, particularly one that provides training. Idmaj in Morocco followed successfully a similar approach with a youth wage subsidy to a lower wedge between the total cost of labor and take-home pay in a context of high taxation of formal labor. Liberia’s Emergency Employment Action Programme (LEEP/LEAP) 60 was another employment program with a strong public works component, in this case addressed in particular to demilitarized youth. Although effective in increasing youth employment in some circumstances, subsidized employment programs can have negative side effects. Cost effective wage subsidy interventions with positive employment effects for beneficiaries have been demonstrated in Poland and the Czech Republic61 and in Argentina,62 but these evaluations may not adequately account for “deadweight loss” (paying an employer for something he or she would have done without the payment) or substitution effects (firing a regular worker so as to be able to hire a subsidized worker). Although they provide employment during the period of public employment, public works programs have not been shown to have a significant effect on employment after the end of the program.63 A positive effect on income has been shown in Argentina64 but not in Ethiopia,65 and Martin Ravallion has 60 http://pdf.usaid.gov/pdf_docs/PNADF305.pdf 61 Betcherman et. al. (2007). 62 Galasso et. al. (2002). 63 Betcherman et. al. (2007). 64 Jalan and Ravallion (1999). 39 argued in several papers66 that alternative policies are likely to be more cost effective once opportunity cost is considered. Public works programs involve direct job creation through public works or other activities that produce public goods or services. They can alleviate unemployment or short-term poverty by creating temporary jobs and can help disadvantaged, poor, and long-term unemployed workers to regain contact with the labor market. Governments can manage these projects directly or contract with non-profit organizations or private businesses. On the positive side, these programs can lead to the production of public goods/services and develop basic physical or social infrastructure; indeed, in many cases, this, rather than job creation, is the main objective. These programs can also be effective short-term safety nets. On the negative side, it is often observed that the long-term labor market impact of these programs is insignificant, and, in some countries, there is a stigma attached to public works jobs which may decrease the employability of participants over the long run. Public works programs for youth can affect outcomes other than employment and income. For example, the Temporary Income Support Program in El Salvador (PATI) aims to guarantee a minimum level of income to poor urban families, while providing a labor market experience at the municipal level. Unlike traditional income support programs, PATI funds the participation of individuals working in projects submitted by municipalities, with emphasis on social services provision, and also includes an innovative training component that aims to enhance beneficiaries’ technical skills and their labor market “soft skills.” Early results looking at the impact added by the life-skills training component show improved labor market outcomes as well as greater self-esteem and greater optimism towards the future. Public works may contribute to social cohesion in fragile environments. These programs are common in many post-conflict countries, where the programs often include former 65 Gilligan et. al. (2008). 66 Ravallion (1999), Ravallion and Datt (1995); Datt and Ravallion (1994). 40 child/youth-soldiers. For example, in Sierra Leone, soon after the conflict ended, a workfare program was launched to help rebuild infrastructure and provide short-term employment opportunities to poor and ex-combatants. Public works programs were also launched and scaled up in Liberia, Yemen, Sierra Leone, Sudan, Guinea, and Guinea Bissau after the conflicts ended. In Nepal, which has just emerged from a decade-long internal conflict, a national public workfare program is being designed. Recent evidence is also suggestive that public works may boost social cohesion and social inclusion. For example, the public works program in Serbia has shown that self-confidence and social inclusion of the participants was strengthened following participation.67 Similarly, the findings from Argentinian Jefes and from the Indian NREGA are also promising.68 A different form of public works, formal or informal public and civic service, provides an opportunity for youth to “play an active role in community and national d evelopment while learning new skills, increasing their employability, and contributing to their overall personal development.”69 Services may include providing basic health services in public health clinics, building sustainable housing, literacy tutoring, protecting the environment, and building small-scale infrastructure, for example. While the evidence differs from program to program, it illustrates that public services can have positive impacts on employability and further education as well as on civic responsibility and self-esteem. The assessments of more than 200 international civil service programs have exhibited an increase in work skills, career options, and education as well as higher self-esteem and sense of civic responsibility.70 While mandatory national programs as an alternative to military service can be expensive and politically manipulated, voluntary 67 The Beautiful Serbia program combines vocational training and public works in the construction sector for disadvantaged unemployed in Serbia and Montenegro in 2004 and 2005. See Bonin and Rinne (2006). They show that the positive impact of this program is much stronger in terms of subjective well-being than of labor market outcomes – the latter is insignificant. 68 See Galasso and Ravallion (2004) for Argentina and Azam (2012) for India. 69 See Cunningham et al. (2008a). 70 See McBride et al. (2003). 41 targeted programs for the vulnerable can be more successful.71 A quasi-experimental evaluation of the national service program Americorps (US) to support community service showed that participants had a greater incidence of post-program civic engagement, more positive attitudes towards employment, and a higher likelihood of public service careers, but no significant increase in educational attainment as compared to a control group.72 Youth Service Canada showed no statistically significant impacts on earnings, though a positive significant impact on post-program education and training. Furthermore, the program was considered valuable by its participants in terms of their personal development, especially regarding self-confidence and knowledge about finding a job.73 In Jamaica’s National Youth Service program 60 percent of participants transition to permanent employment or continue their studies, while only 34 percent of those not participating do so.74 Looking forward, designing more comprehensive packages which combine assistance with other promotional activities seems the most promising approach to foster employability in the short run. Public works are increasingly offered as comprehensive packages in low capacity contexts, combining, for example, a training component or access to credit, such as in the El Salvador PATI program;75 or additional technical and life skills training, such as a youth program in Papua New Guinea; or the Sierra Leone and Liberia Youth Employment Support Programs that provide additional integrated compulsory literacy, numeracy, and life skills training. Public works participants may also be linked with financial services such as microcredit or saving accounts. In Bangladesh, the beneficiaries of the Rural Maintenance 71 See World Bank (2005) and McBride et al. (2003). 72 See Frumkin et al. (2009). 73 See Human Resources Development Canada (1999). 74 See Cunningham et al. (2008a,b). 75 El Salvador Temporary Income Support Program (PATI). PATI guarantees a minimum level of income to poor urban families and provides labor market experience at the municipal level. In contrast to traditional income support programs, PATI funds the participation of individuals working on projects submitted by municipalities, with emphasis on social services provision. It also includes an innovative training component that aims to enhance beneficiaries’ technical skills and their labor market “soft skills.” Th e government expects to target youth between 16 and 24 years of age living in urban areas, as well as female household heads. 42 Program are referred to microfinance institutions.76 Evidence shows that three years after exiting the program, almost 80% of participants were still self-employed in micro-enterprise activities.77 Looking forward, these programs should diversify the work conditionality activities by establishing direct links with activities that are in strong demand by the labor market. 4.5 Labor Market Reform Labor market reforms comprise the most diverse set of youth employment interventions. These reforms aim to remove specific constraints in the macroeconomic or business environment by abrogating or changing legislation or policy that can have a negative effect on employment. For example, introducing a youth sub-minimum wage could allow some young people to find jobs that would previously have been unattainable when the (single) minimum wage was set at a level appropriate for older adults. Alternatively, allowing curfew exemptions for youth returning from work can make it possible for some young people to occupy positions which require them to work late and which could not have previously been held by youth who cannot be outdoors after a specified time. The target group for labor market reforms depends heavily on the legislation being reformed. Although some legislation may affect all youth equally (generalized curfews, for example), others may affect specific sub-groups of youth (rules against men and women working together, for instance). However, since broad labor market reform could affect all individuals in the labor market in question, this type of intervention has the potential to have a larger effect than smaller, often pilot, programs that target a specific population. The effectiveness of the reforms, again, depends on the legislation being reformed as well as the enforcement capacity. Although some rule or piece of legislation may appear to have negative effects on employment, the ability to enforce the legislation is a more important 76 They are required to participate in mandatory savings plans and receive training in numeracy, income- generating skills, and micro-enterprise management. They are referred to microfinance institutions and receive business management advice one year after exiting the program. 77 See Hashemi and Rosenberg (2006). 43 criterion. Some studies have found that raising minimum wages, for example, can increase employment and total earnings due to a “lighthouse effect,” in which individuals who are not working enter employment because they expect a higher wage than is currently being paid, even if it is not as high as the legal minimum wage. 4.6 A Summary of Labor Market Programs for Youth Table 3 shows the anticipated and observed effects for different types of programs in different settings (Low-, Middle- or High-Income, Rural or Urban) and for different beneficiary populations (skilled and unskilled). The anticipated effects are based on economic theory and consider likely outcomes (++ means a large positive expected outcome, + means a small positive outcome expected, and 0 means no expected outcome). The observed effects column summarizes the number of impact evaluations found in YEI- registered programs of a particular type in the given setting that have positive impacts, divided by the total number of impact evaluation studied for that cell. 78 Table 3: Anticipated Effects of Different Programs by Setting and Beneficiary Type 78 The YEI does not collect information on impact evaluation results differentiated by skill level, thus the primary basis for determining how program effects differ by skill level is theory. 44 5. An Agenda Going Forward This section lays out an agenda for research and policy analysis in the area of human development to help both deepen the understanding of youth employment issues and improve the selection, design, and implementation of youth employment programs. The agenda has three pillars: (1) providing better understanding of the causes and consequences of poor labor market outcomes for youth; (2) developing tools to guide the design and implementation of youth employment programs; and (3) supporting a second generation of impact evaluation that focus on assessing how different design features of a given program – including interactions with other programs – affect labor market outcomes for youth. 5.1 Diagnosis –Understanding the causes and consequences of youth unemployment The effective design of youth employment programs requires an understanding of the challenges and constraints facing youth that negatively affect labor market outcomes, and the social implications of these constraints. These are both very complex issues and there is a limit to how much can be done based on available data and resources. It is therefore important to invest in data collection and monitoring, including the design of targeted quantitative and qualitative surveys to better assess the factors that affect employment outcomes at the individual, market, and macro levels. Data collection and monitoring. It is necessary to have a mechanism in place to systematically collect and process information about youth labor market outcomes across countries. Currently there are today several standardized surveys that provide information on labor market conditions (680 surveys covering 150 countries).79 These can be complemented with new data sources such as the Gallup World Poll, an annual survey covering 1,000 individuals per country in roughly 150 countries, which can contribute more timely information on both youth employment outcomes and the emotional well-being of young people. Unfortunately, many countries, particularly those with low income, still do 79 See Montenegro and Hirn (2008). 45 not have regular nationally representative surveys of the labor market. It is important to promote investments in these surveys, by building the institutional capacity of statistical offices and providing technical assistance for survey design and implementation. Improvements in the design of surveys could, for example, come from the introduction of rotating panels, where individuals are followed over time. New analytic work. As new data become available, analysis of that data can deepen the knowledge base on the determinants and consequences of youth employment outcomes. The goal would be to test the link between the three sets of factors discussed in Section 3 and employment outcomes. For instance, little is known about the types of skills that affect youth employment outcomes.80 More generally, there are open questions about the factors that affect transitions into self- or wage employment, determine participation rates, and influence sectoral choices; for instance, demographic factors and fertility rates are examples of non-market sources of employment outcomes. There is also a need to examine job- search strategies and employer behaviors, such as statistical or prejudicial discrimination or lack of information, which may affect hiring decisions. In addition to learning more about the causes of youth unemployment, it is important to better understand the consequences of high unemployment among youths for living standards/well-being, productivity, and social cohesion. As discussed above, the long-term effects of youth unemployment are poorly understood. Several papers have found substantial degrees of scarring in OECD countries,81 but this may be less of an issue in developing countries. Remaining idle or working in a menial job as a youth might have other impacts on human capital accumulation and productivity.82 80 A recent study finds that cohorts that score higher on cognitive exams have significantly less unemployment, controlling for country characteristics (see Lee and Newhouse, 2012). 81 See, for example, Kahn (2010), Oreopoulos and Van Wachter (2006), Bell and Blanchflower (2010). 82 Preliminary evidence from Indonesia has found moderate evidence of scarring, as shocks from 10 years ago were still apparent in decreased male idleness and increased self-employment (Giles, Newhouse, and Witoelar (2011)). On the other hand, analysis from China found that scarring was short-lived, as youth graduating in a poor economy “caught up” in three years [Chi and Freeman (2012)). 46 Employment constraints assessment surveys. Ultimately, generic labor force surveys with limited regional resolution will not be enough to fully understand the constraints facing specific youth population groups in a given geographic region. The proposal is to develop a toolkit to assess individual, market, and macro constraints. This toolkit would combine different instruments: surveys of potential program participants, including modules to measure technical, cognitive and non-cognitive skills; focus groups to gather information on perceptions about employment challenges and constraints in the labor market; and labor market assessments to understand how local and national socio-economic factors affect labor market opportunities for youth. 5.2 Treatment – Designing and implementing effective interventions Eventually, the impact of different interventions to improve youth employment outcomes will depend on how they are designed and implemented. The second part of the youth agenda would need to focus on the development of guidelines to set up alternative programs and policy reforms. Program design. For all programs there are key design decisions that need to be made that can directly affect their performance. In the case of training programs, for instance, it matters whether the delivery is in the classroom or on-the-job. The types of skills that are targeted, the duration of the training, pedagogical methods, and the types of assessments administered are all important elements of a program’s design.83 Similarly in the case of wage subsidies, employers or individuals can be targeted; the aim can be to create new jobs or protect current jobs; and the transfers can take the form of direct payments or reductions in social security contributions. Other issues include the level of the subsidy, its duration, and eligibility conditions (e.g., by type of job, level of the wage bill, and number of workers). Performance also depends on how a given program shares resources and interacts with others. Sharing administrative processes and data not only reduces operational costs but 83 See Honorati et al., (forthcoming). 47 also increases accuracy and reduces leakages. Equally important, several programs need to be implemented as a package in order to have an impact; the composition of the package depends on the targeted group of beneficiaries and the constraints they face. An example is training programs aiming to improve access to wage employment among vulnerable youth, which often need to be accompanied by intermediation services, counseling, and in some cases wage subsidies. Administrative tools. Regarding implementation, all programs discussed in this paper, be they training, counseling, intermediation, or support to the self-employed, share basic “nuts and bolts.” These include beneficiary identification systems, targeting arrangements, registries, monitoring and reporting systems, and contracting and payment systems. These administrative tools reduce the chances of leakages and abuse, thus improving the possibility for meaningful impact from the programs. At the same time, technologies such as biometric data collection are expanding the possibilities available to policymakers. Contracting and payment systems are among the most important, yet difficult, business processes. These systems must track beneficiaries (e.g., for the payment of training allowances), employers (e.g., for the payment of wage subsidies), and providers (for the payment of services delivered). In the case of providers, critical issues include certification and eligibility to participate in competitive biddings (relevant for private companies, NGOs, and/or individuals); contracting arrangements to create accountability and incentives to deliver quality services; and monitoring and enforcement capacity.84 Providers can also be paid directly by beneficiaries via vouchers, but vouchers are only administrative tools that do not dispense with addressing the issues above. In many cases, service providers are currently public agencies managed by civil servants, who are not paid on the basis of services rendered, but rather according to historical budgets. Public providers need not be the default option, and may even be made to adhere to the same rules as private providers when they participate in the implementation of the programs. 84 In the case of training programs, in order to be able to establish minimum quality standards, some countries have adopted National Qualifications Frameworks that certified providers would need to respect. 48 Under this part of the agenda, the goal is to define a typology of design and implementation choices for each type of program, and indicate under which circumstances some choices/interactions should be preferred – based on available knowledge. 5.3 Monitoring, evaluation, and learning Many youth employment programs have been implemented around the world, yet there is still a lack of information about what works and in what situation. The Youth Employment Inventory shows that, of the 500 programs registered, 40 percent only provide basic descriptive information and have not even had a process or performance evaluation. Only 22 percent have had an impact evaluation, and of these only 36 have also had a cost-benefit analysis. Most of the evaluated impacts are positive, but the lack of evidence on the majority of programs, especially of a cost-benefit nature, limits the types of conclusions that can be drawn in terms of design and implementation.85 Table 4: Impact evaluations of youth employment programs (from the YEI) There is a need to adopt guidelines for the integration of monitoring and evaluation systems at the design stage. The M&E strategy should allow for the examination of design features that make a difference while enabling managers to compare the performance of different 85 Please note that 17 programs have ongoing IE and one program (Community Youth Mapping) does not look at earnings or employment as outcome variables (so program impact for Community Youth Mapping on earnings and employment is NA). These 18 programs are not included in the table. 49 components and regions and facilitating effective performance management, if possible down to the level of the individual trainer/service provider. As cost data are often not collected in the same systems as input and output data, the M&E strategy needs to plan on explicitly collecting the data necessary for cost-benefit analyses, both in terms of the internal rate of return for beneficiaries and relative to the overall net benefit or cost of the program for the country. The impact and process evaluations must define and focus on the most important outstanding questions. For example, programs will have heterogeneous effects on different types of workers. In addition, most of the evidence gathered to date has been from relatively small pilot programs. It is necessary to develop a better understanding of spillovers and other “general equilibrium” effects, especially when there is a possibility that the program will be scaled up.86 Well-designed evaluations can identify whether these programs represent effective ways to improve labor market outcomes, promote social inclusion, and reduce youth crime and violence, while identifying which design and implementation factors are associated with greater cost-effectiveness. They can also provide important lessons on how these types of programs are best run at large scale. 6. Conclusions This paper has reviewed the main challenges facing youth in the labor market across countries and the types of interventions that governments have implemented to address them. Accumulated experience suggests that comprehensive programs are more likely to succeed than those that attempt to address only one issue at a time, given that the constraints facing young people are complex and multifaceted. Experience also shows that some programs that are likely to succeed in some areas may not succeed in others, and that context is a determining factor. Where employers demand workers with specific skills and such workers are unavailable, a short-run market failure relative to the supply of skills 86 Progress along these lines can be made by evaluating large-scale public programs that seek to provide public works and/or training to youth, such as the Kenya Youth Employment Program and the Urban Youth Employment Program in Papua New Guinea. 50 clearly exists and may require public intervention. Where there are many people with adequate skills and no demand from employers, the constraints may be elsewhere. The paper has argued that improving labor market outcomes for youth may require targeted interventions. Many of the policies that government implement to facilitate the creation of business and jobs will certainly help youth as well as other workers. But youth might face more constraints relative to adult workers when accessing these jobs – including when they take the form of self-employment opportunities. They have less or no work or job-search experience, they are less connected to jobs through their social networks, and they are less likely to have access to credit to finance a new business. In some countries demographic factors may also work against young jobseekers. Furthermore, the social and private costs of poor labor market outcomes are probably higher for youth than for older workers. Evidence from developed countries shows that prolonged unemployment while young, and problems negotiating the transition to productive employment, can affect workers’ productivity, lifetime earnings, physical and mental health, risk-taking behavior, and longevity. But we are unable to say with certainty that these effects hold among young people in developing countries. The proposed agenda on youth employment has three main elements: diagnostics, treatment, and evaluation. Better diagnostics require better data and tools to understand the factors that determine labor market outcomes among youth and to be able to choose the most effective set of interventions. Better treatment, through improved designs and more effective implement of youth employment programs, requires capturing tacit knowledge among practitioners about successful experiences and extracting operational lessons from ongoing impact evaluations. Finally, better evaluations require that the focus be not on whether a given program works but why, as well as the degree of cost- effectiveness. 51 References Akpan, Sunday Brownson (2010). Encouraging Youth’s Involvement in Agricultural Production and Processing. IFPRI Nigeria Strategy Support Program Policy Note No. 29. Almeida, Rita, Jere Behrman and David Robalino (eds.), 2012. The Right Skills for the Job, (Washington, DC: The World Bank). Azam, Mehtabul, 2012. “The Impact of Indian Job Guarantee Scheme on Labor Market Outcomes: Evidence from a Natural Experiment,” IZA Discussion Paper no. 6548, May. Bell, David, and David Blanchflower, 2010. “Youth Unemployment: Déjà vu?” IZA Discussion Paper no. 4705. Betcherman, G., Godfrey, M., Puerto, S., Rother, F. & Stavreska A. (2007). A Review of Interventions to Support Young Workers: Findings of the Youth Employment Inventory . SP Discussion Paper No. 0715. The World Bank. Bloom, David E. and David Canning, 2005. “Global Demographic Change: Dimensions and Economic Significance,” Cambridge, MA: Harvard Initiative for Global Health Working Paper No.1. Available online at http://www.hsph.harvard.edu/pgda/working/working_paper1.pdf. Bowman, P. J., 1990. “The Adolescent-to-Adult Transition: Discouragement among Jobless Black Youth,” New Directions for Child and Adolescent Development, (46): 87–105. Bryan, Gharad, Shyamal Chaudhury, and Mushfiq Mobarak, 2012. “Seasonal Migration and Risk Aversion,” Centre for Economic Policy Research Working Paper 8739. Chi, Wei, Richard Freeman, and Hongbin Li, 2012. “Adjusting to Really Big Changes: The Labor Market in China 1989-2009,” NBER Working Paper no. 17721. Cho, Yoonyoung and David Newhouse, 2011. “How did the great recession affect different types of workers ? Evidence from 17 middle-income countries,” World Bank Policy Research Working Paper no. 5636, April. 52 Cincotta, Richard P., Robert Engelman, and Daniele Anastasion (2003) The Security Demographic: Population and Civil Conflict After the Cold War. Washington, DC: Population Action International. Cunningham, W., M- Sanchez-Puerta and A. Wuermli, 2010. “Active Labor Market Programs for Youth: A Framework to Guide Youth Employment Interventions,” World Bank Employment Policy Primer, November. Datt, G. & Ravallion, M. (1994). Transfer Benefits from Public-Works Employment: Evidence for Rural India. The Economic Journal, Vol 104, No. 427, pp. 1346–1369. Galasso, E., Ravallion, M., & Silvia, A. (2002). Assisting the Transition from Workfare to Work: A Randomized Experiment. Policy Research Working Paper. World Bank. Giles, John T, David Newhouse, and Firman Witoelar, 2012. "Stuck for Life? The Long-Term Consequences of Initial Informality in Indonesia," World Bank. Washington, D.C. Gilligan, D. O., Hoddinott, J. & Taffesse, A. S. (2008). The Impact of Ethiopia’s Productive Safety Net Program and its Linkages. IFPRI Discussion Paper 00839. Gregg, P., and E. Tominey, 2005. “The Wage Scar from Male Youth Unemployment,” Labour Economics 12 (4): 487–509. Groh, Matthew, Nandini Krishnan, David McKenzie and Tara Vishwanath, 2012. “Soft Skills or Hard Cash? The impact of training and wage subsidy programs on female youth employment in Jordan,” presented in the World Bank’s DEC Poverty and Applied Micro Seminar Series, June 6, Washington DC. Heckman, J. J., J. Stixrud and S. Urzua, 2006. “The Effects of Cognitive and Noncognitive Abilities on Labor Market Outcomes and Social Behavior,” Journal of Labor Economics, vol. 24, pp. 411-482. International Labor Organization (ILO), 2011. “Youth Employment through Local Economic 53 Development in Quang Nam Province.” ILO Evaluation Series, ILO. Jalan, J. & Ravallion, M. (1999). Income Gains to the Poor from Workfare: Estimates for Argentina‘s Trabajar Program. Policy Research Working Paper. World Bank. Jensen, Robert, 2010. “The (Perceived) Returns to Education and the Demand for Schooling,” Quarterly Journal of Economics, vol. 125, no.2, pp. 515-548. Jensen, Robert, 2012. “Do Labor Market Opportunities Affect Young Women's Work and Family Decisions? Experimental Evidence from India,” Quarterly Journal of Economics, vol 127, no.2, pp. 753-792. Kahn, Lisa B, 2010. “The Long-term Labor Market Consequences of Graduating College in a Bad Economy,” Labour Economics, vol. 17 no. 2, pp. 303-316. Lee, Jean, and David Newhouse, 2012. "Cognitive Skills and Labor Market Outcomes," Background paper for the World Development Report 2013. Lehr, David, 2008. “Microfranchising at the Base of the Pyramid,” Working Paper, the Acumen Fund. Mahmood, Moazam, David Newhouse, David A. Robalino, Friederike Rother, and Catherine Saget, 2012. “Inventory of Policy Responses to the Financial and Economic Crisis,” ILO/World Bank Joint Synthesis Report. Margolis, David N., Jean A. Lee, David Newhouse, David A. Robalino and Eduardo Zylberstajn, forthcoming. “A Framework for Understanding Self -Employment and Small- Scale Entrepreneurship” IZA Journal of Labor Policy. Mason, A., 2005. “Demographic Transition and Demographic Dividends in Developed and Developing Countries,” United Nations Expert Group Meeting on Social and Economic Implications of Changing Population Age Structures, Mexico City. 54 Montenegro, C. and M. Hirn, 2008. “A New Disaggregated Set of Labor Marke t Indicators Using Standardized Household Surveys from Around the World,” World Development Report 2009 Background Paper. Montenegro, C. and C. Pages, 2003. “Who Benefits from Labor Market Regulations? Chile 1960-1998,” NBER Working Paper no. 9850. Oreopoulos, Philip, Till von Wachter, and Andrew Heisz, 2006. “Hysteresis and Heterogeneity in the Market for College Graduates,” NBER Working Paper no. 12159. Ravallion, M. (1999). Appraising Worfare. The World Bank Research Observer, Vol. 14, No. 1, pp. 31–48. Ravallion, M. & Datt, G. (1995). Is Targeting through a Work Requirement Efficient? Some Evidence for Rural India. In Van de Valle, D. & Nead, K., Public Spending and the Poor: Theory and Evidence. Baltimore: The Johns Hopkins University Press. van Lieshout, Susanne; Sievers, Merten; Aliyev, Mirza (2012). Start and improve your business : global tracer study 2011: ILO’s business management training programme. International Labour Office, Geneva: ILO. World Bank, (2012). “What Makes Entrepreneurship Education and Training Programs Effective”, Concept Note, Reviewed May 2, 2012. World Bank (2012b). Can skills training programs increase employment for young women? The Case of Liberia. PREM Gender Results Series, October. Youth Employment Inventory at: http://www.youth-employment-inventory.org/ 55 Social Protection & Labor Discussion Paper Series Titles 2011-2013 No. Title 1308 Youth Employment: A Human Development Agenda for the Next Decade by David Robalino, David Margolis, Friederike Rother, David Newhouse and Mattias Lundberg, June 2013 1307 Eligibility Thresholds for Minimum Living Guarantee Programs: International Practices and Implications for China by Nithin Umapathi, Dewen Wang and Philip O’Keefe, November 2013 1306 Tailoring Social Protection to Small Island Developing States: Lessons Learned from the Caribbean by Asha Williams, Timothy Cheston, Aline Coudouela and Ludovic Subran, August 2013 1305 Improving Payment Mechanisms in Cash-Based Safety Net Programs by Carlo del Ninno, Kalanidhi Subbarao, Annika Kjellgren and Rodrigo Quintana, August 2013 1304 The Nuts and Bolts of Designing and Implementing Training Programs in Developing Countries by Maddalena Honorati and Thomas P. McArdle, June 2013 1303 Designing and Implementing Unemployment Benefit Systems in Middle and Low Income Countries: Key Choices between Insurance and Savings Accounts by David A. Robalino and Michael Weber, May 2013 1302 Entrepreneurship Programs in Developing Countries: A Meta Regression Analysis by Yoonyoung Cho and Maddalena Honorati, April 2013 1301 Skilled Labor Flows: Lessons from the European Union by Martin Kahanec, February 2013 1220 Evaluating the Efficacy of Mass Media and Social Marketing Campaigns in Changing Consumer Financial Behavior by Florentina Mulaj and William Jack, November 2012 1219 Do Social Benefits Respond to Crises? Evidence from Europe & Central Asia During the Global Crisis by Aylin Isik-Dikmelik, November 2012 1218 Building Results Frameworks for Safety Nets Projects by Gloria M. Rubio, October 2012 1217 Pension Coverage in Latin America: Trends and Determinants by Rafael Rofman and Maria Laura Oliveri, June 2012 1216 Cash for Work in Sierra Leone: A Case Study on the Design and Implementation of a Safety Net in Response to a Crisis by Colin Andrews, Mirey Ovadiya, Christophe Ribes Ros and Quentin Wodon, November 2012 1215 Public Employment Services, and Activation Policies by Arvo Kuddo, May 2012 1214 Private Pension Systems: Cross-Country Investment Performance by Alberto R. Musalem and Ricardo Pasquini, May 2012 1213 Global Pension Systems and Their Reform: Worldwide Drivers, Trends, and Challenges by Robert Holzmann, May 2012 1212 Towards Smarter Worker Protection Systems: Improving Labor Regulations and Social Insurance Systems while Creating (Good) Jobs by David A. Robalino, Michael Weber, Arvo Kuddo, Friederike Rother, Aleksandra Posarac and Kwabena Otoo 1211 International Patterns of Pension Provision II: A Worldwide Overview of Facts and Figures by Montserrat Pallares-Miralles, Carolina Romero and Edward Whitehouse, June 2012 1210 Climate-Responsive Social Protection by Anne T. Kuriakose, Rasmus Heltberg, William Wiseman, Cecilia Costella, Rachel Cipryk and Sabine Cornelius, March 2012 1209 Social Protection in Low Income Countries and Fragile Situations: Challenges and Future Directions by Colin Andrews, Maitreyi Das, John Elder, Mirey Ovadiya and Giuseppe Zampaglione, March 2012 1208 World Bank Support for Pensions and Social Security by Mark Dorfman and Robert Palacios, March 2012 1207 Labor Markets in Middle and Low Income Countries: Trends and Implications for Social Protection and Labor Policies by Yoonyoung Cho, David Margolis, David Newhouse and David Robalino, March 2012 1206 Rules, Roles and Controls: Governance in Social Protection with an Application to Social Assistance by Lucy Bassett, Sara Giannozzi, Lucian Pop and Dena Ringold, March 2012 1205 Crisis Response in Social Protection by Federica Marzo and Hideki Mori, March 2012 1204 Improving Access to Jobs and Earnings Opportunities: The Role of Activation and Graduation Policies in Developing Countries by Rita Almeida, Juliana Arbelaez, Maddalena Honorati, Arvo Kuddo, Tanja Lohmann, Mirey Ovadiya, Lucian Pop, Maria Laura Sanchez Puerta and Michael Weber, March 2012 1203 Productive Role of Safety Nets by Harold Alderman and Ruslan Yemtsov, March 2012 1202 Building Social Protection and Labor Systems: Concepts and Operational Implications by David A. Robalino, Laura Rawlings and Ian Walker, March 2012 1201 MicroDeterminants of Informal Employment in the Middle East and North Africa Region by Diego F. Angel-Urdinola and Kimie Tanabe, January 2012 1120 Employment Generation in Rural Africa: Mid-term Results from an Experimental Evaluation of the Youth Opportunities Program in Northern Uganda by Christopher Blattman, Nathan Fiala and Sebastian Martinez, December 2011 1119 Measuring Governance and Service Delivery in Safety Net Programs by Gloria M. Rubio, September 2011 1118 Assessing Safety Net Readiness in Response to Food Price Volatility by Margaret Grosh, Colin Andrews, Rodrigo Quintana, Claudia Rodriguez-Alas, September 2011 1117 Social Safety Nets in Fragile States: A Community-Based School Feeding Program in Togo, August 2011 (also available in French) by Colin Andrews, Elena Galliano, Carolyn Turk and Giuseppe Zampaglione, August 2011 1116 Strengthening Governance of Social Safety Nets in East Asia by Sara Giannozzi and Asmeen Khan, August 2011 1115 International Portability of Health-Cost Coverage: Concepts and Experience by Martin Werding and Stuart McLennan, July 2011 1114 Liberia’s Cash For Work Temporary Employment Project: Responding to Crisis in Low Income, Fragile Countries by Colin Andrews, Prospère Backiny-Yetna, Emily Garin, Emily Weedon, Quentin Wodon and Giuseppe Zampaglione, July 2011 1113 Employability and Productivity among Older Workers: A Policy Framework and Evidence from Latin America by Edmundo Murrugarra, July 2011 1112 Cash Transfers, Children and the Crisis: Protecting Current and Future Investments by Ariel Fiszbein, Dena Ringold, Santhosh Srinivasan, June 2011 1111 Severance Pay Programs around the World: History, Rationale, Status, and Reforms by Robert Holzmann, Yann Pouget, Milan Vodopivec and Michael Weber, May 2011 1110 Portability of Pension, Health, and other Social Benefits: Facts, Concepts, Issues by Robert Holzmann and Johannes Koettl, May 2011 1109 Disability and Poverty in Developing Countries: A Snapshot from the World Health Survey by Sophie Mitra, Aleksandra Posarac and Brandon Vick, April 2011 1108 Advancing Adult Learning in Eastern Europe and Central Asia by Christian Bodewig and Sarojini Hirshleifer, April 2011 1107 Results Readiness in Social Protection & Labor Operations by Laura Rawlings, Maddalena Honorati, Gloria Rubio and Julie Van Domelen, February 2011 1106 Results Readiness in Social Protection & Labor Operations: Technical Guidance Notes for Social Service Delivery Projects by Julie Van Domelen, February 2011 1105 Results Readiness in Social Protection & Labor Operations: Technical Guidance Notes for Social Safety Nets Task Teams by Gloria Rubio, February 2011 1104 Results Readiness in Social Protection & Labor Operations: Technical Guidance Notes for Social Funds Task Teams by Julie Van Domelen, February 2011 1103 Results Readiness in Social Protection & Labor Operations: Technical Guidance Notes for Labor Markets Task Teams by Maddalena Honorati, February 2011 1102 Natural Disasters: What is the Role for Social Safety Nets? by Larissa Pelham, Edward Clay and Tim Braunholz, February 2011 1101 North-South Knowledge Sharing on Incentive-based Conditional Cash Transfer Programs by Lawrence Aber and Laura B. Rawlings, January 2011 To view Social Protection Discussion papers published prior to 2011, please visit www.worldbank.org/sp Abstract This paper reviews the main challenges facing countries in attempting to improve labor market outcomes among youth, focusing on the issues that became starkly visible during the recent financial crisis. In order to better identify and set up human development interventions, the paper proposes an agenda that focuses on three areas: (1) improving the understanding of the causes and consequences of poor labor market outcomes for youth; (2) continuing to learn from the evaluation of pilots and programs that aim to promote productive employment among young people; and (3) addressing implementation issues which frequently overwhelm the best designs. The paper utilizes research on youth employment to take stock of youth labor market outcomes across regions, focusing on inactivity, unemployment, and employment indicators. A review of what is known about current interventions, including those that appear in the Youth Employment Inventory database of programs, provides the basis for determining the efficacy of five categories of intervention: (i) skills training (including vocational training, on-the-job-training programs, literacy and numeracy programs, second-chance and equivalency programs, and soft-skills programs); (ii) entrepreneurship promotion (financial assistance, technical assistance, and entrepreneurship training); (iii) subsidized employment (including wage subsidy programs, public works, and public/community service programs); (iv) employment services (including search assistance and access to labor market information, job counseling and placement services, and financial assistance for job search); and (v) reforms to labor market regulation (including anti-discrimination legislation) training programs, wage subsidies. Finally, the paper proposes an agenda for research and policy analysis in the area of human development that is expected to help both deepen the understanding of youth employment issues and improve the selection, design, and implementation of youth employment programs. About this series... Social Protection & Labor Discussion Papers are published to communicate the results of The World Bank’s work to the development community with the least possible delay. The typescript manuscript of this paper therefore has not been prepared in accordance with the procedures appropriate to formally edited texts. The findings, interpretations, and conclusions expressed herein are those of the author(s), and do not necessarily reflect the views of the International Bank for Reconstruction and Development / The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The author(s) attest(s) that the paper represents original work. It fully references and describes all relevant prior work on the same subject. For more information, please contact the Social Protection Advisory Service, The World Bank, 1818 H Street, N.W., Room G7-803, Washington, DC 20433 USA. Telephone: (202) 458-5267, Fax: (202) 614-0471, E-mail: socialprotection@worldbank.org or visit us on-line at www.worldbank.org/sp. © 2013 International Bank for Reconstruction and Development / The World Bank Printed on recycled paper