SOCIAL PROTECTION DISCUSSION PAPER No. 2511 | MARCH 2025 State of Social Protection Report 2025 The 2-Billion-Person Challenge Background Paper #4 Optimizing Labor Market Programs and Strengthening Delivery Systems for Impact and Scale Eliana Carranza Matteo Morgandi Diana Sverdlin © 2025 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: +1 (202) 473 1000; Internet: www.worldbank.org. This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. RIGHTS AND PERMISSIONS The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: +1 (202) 522 2625; e-mail: pubrights@worldbank.org. Abstract This paper examines the design and implementation of labor market (LM) programs to address employment challenges across diverse economic contexts. LM programs are critical in enhancing employment opportunities, increasing productivity, and fostering resilience among workers. The analysis highlights gaps in program spending and its alignment with labor market needs, particularly in low- and middle-income countries (LMICs). Using global data, the paper emphasizes the importance of tailored interventions, robust delivery systems, and strategic investments to address structural constraints and adapt to crises, such as the COVID- 19 pandemic. The paper also includes a framework for aligning LM programs with labor market realities to support inclusive and sustainable economic growth. JEL: Labor market programs, employment policies, public expenditure Keywords: J08, J48, H53 2 Acknowledgement This paper was authored by Eliana Carranza, Matteo Morgandi, and Diana Sverdlin Lisker of the Labor and Skills Global Solutions Group (GSG), Social Protection and Labor (SPL) Global Unit, World Bank, with data analysis support from Alina Zoe Runk, Samira Salwan, as well as Ingrid Mujica and Belén Fontanez from the ASPIRE team. It was prepared as part of the flagship report The State of Social Protection 2025: The 2-Billion-Person Challenge (World Bank 2025). The Atlas of Social Protection: Indicators of Resilience and Equity (ASPIRE) served as the primary data source. The authors are grateful to Joana Silva, Ruth Hill, and Harry Edmund Moroz for their thoughtful peer review. They also acknowledge the administrative support of Agnes Nderakindo Mganga and Angela Maria Rubio. This work was conducted under the supervision of Michal Rutkowski in its initial stages and later under Iffath Sharif, both serving as Global Directors for Social Protection and Labor. The team also benefited from the strategic direction of Jamele Rigolini, Senior Advisor for Social Protection and Labor, and Loli Arribas-Banos, Practice Manager for the Global Engagement Unit. 3 Table of Contents Introduction............................................................................................................................................ 5 I. Labor Market Challenges and Barriers Differ across Countries According to Their Stage of Development and Structural Transformation ........................................................................................ 6 II. Governments Have an Array of Policy Tools and Programs to Intervene in Labor markets ............ 11 III. Successful Labor Market Programs Are Context Specific but Can Adapt to a Changing Economic Landscape ............................................................................................................................................. 14 IV. The Current State of Spending on Labor Market Programs Does Not Fully Address the Constraints Faced .................................................................................................................................................... 20 V. Strong Systems Enable Rapid Responses during Times of Crisis, as Evidenced during the COVID-19 Pandemic .............................................................................................................................................. 31 VI. Labor Market Programs Going Forward.......................................................................................... 35 References ............................................................................................................................................ 38 Table of Figures Figure 1.1: Women’s Labor Force Participation Is Lower Than Men’s Worldwide, and Self- Employment Is More Prevalent among Women in LICs and LMICs ....................................................... 7 Figure 1.2: Youth Face Challenges Everywhere in the World to Become Part of the Productive Workforce ............................................................................................................................................... 8 Figure 1.3: Productivity Increases with Rising Income Levels ................................................................ 9 Figure 1.4: Labor Market Characteristics and Constraints Vary by Country Income Level .................. 11 Figure 2.1: Instruments the Government Can Use to Address Market Failures in Labor Markets ...... 12 Figure 3.1: Conceptual Framework: Programs Should Match Key Labor market Constraints and Adapt Over Time ............................................................................................................................................. 15 Figure 4.1: The Number of Programs by Country Type and Year......................................................... 21 Figure 4.2: Spending on Labor Market Programs is Only Correlated with Unemployment in UMICs . 24 Figure 4.3: Spending on LM Programs Is Correlated with Self-Employment in LICs and LMICs .......... 26 Figure 4.4: Countries at Higher Levels of Income Offer a Wider Range of Labor market Programs, but Spending on Programs Does Not Align with Key Labor Market Constraints........................................ 30 Figure 4.5: Spending per Beneficiary Increases with Income Level ..................................................... 31 Figure 5.1: UMICs and HICs Expanded and Created New LM Programs During COVID-19 While LICs and LMICs Maintained Existing Programs ............................................................................................ 32 Figure 5.2: Increased Spending Occurred Primarily in Existing Program Types and Decreased After the COVID-19 Pandemic ....................................................................................................................... 34 4 Introduction Governments use labor market programs as a tool to stimulate labor demand, enhance the quality of employment, increase the earnings of the self-employed, match workers with opportunities, and respond effectively to the evolving demands and prospects of the dynamic labor market. When effectively implemented, these programs serve a dual purpose in the broader social protection system: providing economic opportunities and fostering resilience within households (World Bank 2022a). Labor market programs can create opportunities by providing recipients, particularly the most disadvantaged, with the information that they need to optimize their skills and capital investments, access markets, and find appropriate jobs. They can also help ensure that governments make investments with significant externalities, be it in skills that are not firm- specific or in programs that support employment for women, youth, or disadvantaged groups, in which the societal gains go beyond the benefits to firms, individuals, and their families (Palacios and Robalino 2020; World Bank 2012). Additionally, labor market programs can play a critical role in making workers more resilient during times of crisis—as evidenced during the COVID-19 pandemic—and during individual- level shocks. Temporary wage subsidies to employers, for example, were widely deployed in 2020–2021 to preserve existing job matches that would have been costly to break and re- form once the crisis subsided. Breaking existing matches is especially costly when the employer-employee match is a source of productivity gains; that is when returns (and costs) to on-the-job learning are high. Similarly, unemployment benefits were made more generous and accessible in the same period to support the many displaced, including workers in nonstandard contracts. In addition to addressing the challenges posed by the COVID-19 crisis, labor market programs can help reskill workers affected by climate change, as recurring environmental shocks alter local job opportunities, or equip individuals with the right skills needed to access global markets, as seen in the case of digital skills. 5 We begin by describing labor market characteristics across countries at different income levels. Section II provides a typology of labor market programs, along with evidence on their effectiveness when applied in the proper context. Section III lays out a conceptual framework that explains why labor market programs should be context specific, in line with the findings of the forthcoming Jobs for Development flagship report. Different countries have different labor market structures which, in turn, call for different program emphases. When countries are grouped according to their income levels, common characteristics and constraints that hinder optimal employment matches become apparent. The framework suggests the types of labor programs that better align with different labor market challenges. Section IV uses the Atlas of Social Protection: Indicators of Resilience and Equity (ASPIRE) data set to describe the current provision of labor market programs across countries. It first looks at the correlation between program spending and the size of the challenge (that is, unemployment and self- employment) and then contrasts the emphasis on specific programs with what the framework would suggest. Section V looks at how countries used their labor market programs and existing delivery systems to respond to COVID-19. Finally, Section 6 concludes. I. Labor Market Challenges and Barriers Differ across Countries According to Their Stage of Development and Structural Transformation Aggregate labor market indicators show minimal variation across countries; yet they mask differences in participation rates by gender and age, types of employment, and labor productivity. Labor force participation rates for countries across income groups and regions hover at around 60 percent, with unemployment rates universally low. However, closer examination reveals significant differences concerning the gender, age, productivity, organizational, spatial, and sectoral composition of employment across income levels. Male labor force participation remains consistent worldwide, but female participation rates exhibit a U-shaped pattern, with the lowest rates found in lower-middle-income countries (LMICs). Men’s labor force participation rates oscillate around 70 percent, with no discernable differences across countries of different income levels. Women’s participation 6 rates are lower than men’s worldwide and exhibit a U-shaped in relation to gross domestic product (GDP). 1 That is, women’s participation rates are higher in low-income countries (LICs) and high- income countries (HICs) and lower in middle-income countries (MICs). Importantly, these disparities are attributed to the type of employment in countries at different income levels and to historical differences in economic structures. 2 Figure 1.1: Women’s Labor Force Participation Is Lower Than Men’s Worldwide, and Self-Employment Is More Prevalent among Women in LICs and LMICs Panel A Panel B Source: World Bank 2022b In LMICs, most workers are self-employed or underemployed. Among men in LICs, the self- employment rate reaches 77 percent, compared to only 14 percent in HICs. For women, the gap is even wider, with 87 percent self-employed in LICs versus only 9 percent in HICs. This 1 The U-shaped trend in female labor force participation has been identified in studies conducted both within and across different countries (Boserup 1970; Goldin 1995; Heath and Jayachandran 2017). 2 In the early stages of development, women often engage in unpaid work within family farms and businesses. As economies grow, economic activity shifts to factories and firms, leading to a decline in female labor force participation. This decline may stem from reduced household reliance on dual incomes, societal biases against women working outside the home, and the challenge of balancing outside employment with childcare responsibilities that tend to fall on women. However, as development progresses, female employment rates rise again, driven by a combination of factors, such as a decline in fertility, which leads to more free time, increased educational attainment, and a shift toward activities that are more intellectually demanding and less physically intensive. Moreover, historical differences in economic structure are also important determinants of female labor force participation levels (Klasen 2019). 7 high prevalence of self-employment reflects the limited availability of attractive wage- employment opportunities, forcing many individuals to create their own jobs. Even those formally employed often find themselves in ‘low-quality’ jobs, forced to accept the first available opportunity regardless of its suitability. This is often exacerbated by a lack of savings or unemployment insurance, leaving individuals financially vulnerable during periods of job search or transition. Underemployment is another significant issue in LICs, evident in the number of hours worked. In LMICs, over 40 percent of employed individuals work less than 35 hours per week (Merotto et al. 2018). Figure 1.2: Youth Face Challenges Everywhere in the World to Become Part of the Productive Workforce Panel A Panel B Source: World Bank 2022b Quality employment remains a persistent challenge for youth across countries at varying income levels. Despite progress in school access and attainment, youth still face higher unemployment than the general working-age population. Globally, more than 13 percent of youth ages 15–24 are unemployed, contrasting with 4 percent of those ages 25 and older. In LICs, 83 percent of the youth workforce turn to self-employment as a last resort, slightly higher than the 80 percent among adults. In both LICs and LMICs, over 60 percent of the youth workforce resort to self-employment as a last option, compared to less than 30 percent of youth in UMICs and UICs. This imposes significant costs on both youth and their societies, 8 as joblessness not only affects immediate income but also creates lasting scars on earnings and employment prospects. While some countries are undergoing a youth bulge, others are experiencing a rapidly aging population, resulting in a declining labor force. Since 2010, Sub-Saharan Africa has seen its youth population increase by over 30 percent, with projections indicating a further 40 percent rise over the next decade (UNDESA 2019). Conversely, Europe and Central Asia, East Asia, and Latin America are witnessing aging populations, a trend that is expected to accelerate. For example, Latin American countries are projected to reach old-age dependency ratios comparable to Europe’s in just 25 years—a transformation that took Europe over 70 years (Rofman and Apella 2020). These demographic shifts have diverse implications. First, in UMICs and HICs, an aging population leads to a shrinking labor force and shortages in various economic sectors. Meanwhile, in LICs and MICs, the lack of quality jobs and their inability to absorb the growing workforce is likely to intensify migratory pressures between countries. Second, the growing elderly population strains pension systems, unless countries can extend the productive life of the workforce or boost productivity. Achieving this potential requires implementing numerous policies to support healthy aging. Figure 1.3: Productivity Increases with Rising Income Levels Source: ILOSTAT 2022. Note: PPP = Purchasing power parity. 9 Labor productivity and consequently earnings vary significantly across country income groups, with output per hour worked approximately 15 times higher in HICs than in LICs. In LICs, output per hour worked averages at US$3.0, whereas in HICs it reaches US$57.80 per hour worked.3 Moreover, there are large and persistent gaps in labor productivity across different types of jobs, demographics, sectors, and locations in developing countries. In many cases, these gaps reflect differences in total factor productivity (Prescott 1998; Romer 1993), yet they can also stem from differences in skill levels or barriers preventing the movement of workers into more productive jobs (McMillan and Rodrik 2011). Labor market programs can help dismantle these barriers and steer countries toward a more productive and jobs-rich path. As economies develop, labor productivity increases through the structural transformation process across sectoral, spatial, occupational, and organizational dimensions. In the early stages of a country’s development, there is a shift away from agriculture toward job creation in the manufacturing and service sectors. Concurrently, development is marked by a spatial transition from rural to urban employment, as more jobs are created in cities. This shift is also accompanied by a change in the occupational composition of employment, from simpler roles that require elementary and general skills to more complex jobs demanding advanced and specific skills. Finally, labor markets undergo an organizational transformation in which workers shift from mostly informal self-employment with little capital to predominantly formal employment in well-capitalized, modern firms. Depending on the stage of transformation, the prevalence of various demand- and supply- side barriers, along with labor market frictions, varies by country income level. For countries at lower levels of income, low demand for wage workers in urban cities, coupled with low productivity for those remaining in rural areas, poses a significant constraint. As income levels rise, inadequate skill supply emerges as a challenge, with 23 percent of firms at lower levels 3 This measure of productivity refers to the total volume of output (measured in terms of GDP) produced per unit of labor (measured in terms of number of employed persons or hours worked) during a given time reference period. 10 of income citing workforce skills as a significant constraint to their operations. In some African and Latin American countries, this share rises to 40–60 percent (World Bank 2023). Information friction—for firms to discern between workers and for workers to discern between job opportunities—represents an important constraint in MICs. Institutional and contextual factors can further compound these challenges. Meanwhile, in HICs, an aging population has led to a surplus of unfilled jobs. Figure 1.4: Labor Market Characteristics and Constraints Vary by Country Income Level Source: Adapted from World Bank (forthcoming). II. Governments Have an Array of Policy Tools and Programs to Intervene in Labor markets Governments can alleviate the most prevalent labor market barriers by intervening through various programs. Labor market programs are usually categorized into three groups: (1) active labor market programs (ALMPs), which require ‘active’ participation from beneficiaries and focus on enhancing the human capital of the workforce (such as vocational training programs), increasing firms’ demand for labor (such as wage subsidies and apprenticeship programs), and assisting in job matching (such as public employment services); (2) unemployment insurance, also known as passive labor market programs, which aim to protect workers during periods of joblessness; and (3) economic inclusion (EI) programs, which combine a range of measures aimed at increasing the earnings and assets of extremely poor and vulnerable households, often feature a strong activation component. 11 Figure 2.1: Instruments the Government Can Use to Address Market Failures in Labor Markets •Supply-side: Programs to •Protection of workers •A multipronged approach ALMPs EI Programs Unemployment Insurance develop the human capital of the during periods of that combines several workforce (such as skills training joblessness or job search types of antipoverty or entrepreneurship support) measures with the explicit •Demand-side: Programs to goal of increasing earnings sustain and expand labor and assests of demand (such as wage subsidies) marginalized groups •Intermediation: Employment services programs to match supply and demand of labor (such as job-search and job- match assistance and information services) Note: Economic inclusion programs are a type of comprehensive active labor market program that tackles supply, demand, and intermediation issues. We depict it as a separate type of instrument because ASPIRE does not currently track these programs under ALMPs. ALMPs, if well-designed and implemented to address labor market constraints, hold the potential to yield successful outcomes. They can incentivize job seeking, improve job readiness, assist in finding suitable employment, and expand job opportunities. Historically, these programs have suffered from underfunding, limited scale, and fragmentation, which understandably limited their effectiveness. Additionally, coordinated efforts from labor market and social institutions and policies are crucial for their success. Early assessments of ALMPs revealed relatively modest effects (McKenzie 2017). However, advancements in the design, targeting, and implementation of programs in developing countries suggest that ALMPs can play a crucial role (Carranza and McKenzie 2024). Unemployment insurance enables jobless workers to prolong their job search, resulting in improved employment outcomes. Research suggests that unemployment insurance reduces the likelihood of workers settling for low-paying or undesirable jobs out of desperation, benefitting both workers and employers (Rothstein 2011; Song et al. 2011). In HICs with ample formal-sector jobs offering good wages, the financial buffer provided by unemployment insurance allows individuals to be more selective in their job searches and wait for opportunities that better match their skills and preferences (Nekoei and Weber 12 2017). Meanwhile, in countries with high levels of informality, unemployment insurance can deter transitions into the informal sector during periods of joblessness (Liepmann and Pignatti 2024). EI programs have demonstrated significant effects on business revenues and asset accumulation, extending beyond their impacts on food security, income, and consumption (Banerjee et al. 2021). These programs are intended to enhance households’ capacity to adapt to and recover from shocks by facilitating access to regular and predictable cash transfers. However, unlike traditional cash transfer programs, EI programs also foster asset accumulation, income diversification, and access to financial services to build resilience. To support productive employment, the design of EI programs is influenced by the characteristics of the local economy and target populations. For instance, in LICs, particularly in rural areas, the emphasis lies on income diversification and transitioning from unpaid or casual work to self-employment. Conversely, in MICs, the focus shifts toward the urban sector and opportunities for wage employment. The choice and size of policy programs must then align with the specific labor market context and address the distinct constraints countries face. To optimize employment outcomes, the selection and emphasis on different interventions—from a menu comprising unemployment insurance, active labor market interventions, and economic inclusion interventions—should be guided by a country-specific diagnosis of labor market characteristics and the most binding constraints, whether on the demand or supply sides of the labor market or in the intermediation between them. In most countries, a key challenge lies in the inadequate demand for workers from the private sector, as well as limited access to capital and technology to engage in productive self-employment. Additionally, supply-side constraints vary by country and encompass inadequate skills, limited job experience, and, especially for women, barriers to participation such as caregiving responsibilities, mobility restrictions, or entrenched social norms. Labor market frictions—such as information gaps regarding job vacancies, ineffective signals of skills and ability, segmented social networks, 13 spatial segregation, and discrimination—can further prevent job matches. Institutional factors may exacerbate these challenges. III. Successful Labor Market Programs Are Context Specific but Can Adapt to a Changing Economic Landscape The effectiveness of labor market programs depends critically on their design as well as the readiness of the institutions tasked with their delivery. Ensuring a strategic blend that optimizes impact and responsiveness is pivotal. Given the unique and multifaceted nature of labor market challenges, customized and comprehensive approaches tend to outperform interventions targeting specific constraints in isolation.4 At times, the context remains static, shaped by structural labor market characteristics and constraints, like those described in the framework outlined in Figure 3.1. However, labor market programs must also show agility to respond to temporal shocks, swiftly adapting to evolving economic conditions and emerging challenges. Crucially, while seizing short-term market opportunities and adapting swiftly to evolving conditions, effective programs should also support (or at least not counteract) medium-term policies for structural transformation. Governments need to assess the existing labor market landscape to create the right mix of programs. Each type of policy tackles a different set of problems and has a different rationale for government intervention. In LICs, ALMPs may involve valuable initiatives such as public works projects, EI programs to facilitate the entry of marginalized groups into the workforce, improvements in the quality of self-employment, and fostering connections with value chains. In LMICs with more diverse labor demands, the focus may shift toward equipping workers with relevant skills and facilitating their integration into private sector roles. In UMICs, ALMPs should continue with skills training, both within the technical and vocational 4 For example, the US$150 million World Bank’s ‘Kenya Youth and Opportunities Project’ was intended to boost employment and earnings for youth who experienced extended unemployment or engaged in vulnerable jobs. Over 300,000 youth received a mix of training and internships in the crafts sector, business start-up grants, coaching, and access to an online one-stop-shop information platform. Those previously employed saw a 20 percent increase in their income, and 48 percent of business support recipients created jobs for others. 14 system and on-the-job training to acquire job-specific human capital, as well as wage subsidies to improve private sector demand. Unemployment insurance schemes should be in place to avoid workers switching to the informal sector. In many HICs, apart from the programs mentioned, efforts should be made to improve public employment services that facilitate matches between firms and workers. Unemployment insurance should be in place to support workers’ smooth consumption during periods of joblessness and facilitate shorter and better employment transitions. While it is critical to tailor programs to the needs of countries at different income levels, maintaining a broad portfolio of programs enhances the labor market systems’ adaptability, enabling expansion and refinement in response to emerging challenges. Figure 3.1: Conceptual Framework: Programs Should Match Key Labor market Constraints and Adapt Over Time Source: Own elaboration. ALMPs include a wide range of interventions aimed at tackling different market failures, such as suboptimal investment in human capital, information frictions, and lack of access to finance, among others. ALMPs have two primary objectives: first, to increase the likelihood of unemployed individuals finding employment, and second, to foster greater inclusion and participation in productive employment. The first objective is typically prioritized by HICs aiming to reduce the duration that individuals remain unemployed and reliant on government benefits. Meanwhile, the second objective holds greater importance 15 in developing countries, where market failures are thought to be larger and prevent access to quality jobs. Given the variety of programs available under ALMPs and their different objectives, all countries stand to benefit from their implementation. Unemployment insurance fosters labor market flexibility by enabling workers to pursue job opportunities that better align with their skills and preferences rather than accepting the first available option out of financial necessity. The rationale behind government intervention is that private insurers are unwilling to offer unemployment insurance because it would predominantly attract those workers at higher risk of losing their jobs, making insurance too costly to provide. Mandating contributions from wages enables the government to pool resources and provide insurance to smooth consumption during periods of joblessness. While desirable in all countries, unemployment insurance works best when (1) there are enough people in formal wage employment to pool resources, and (2) effective monitoring can be conducted. For these reasons, unemployment insurance should primarily be implemented in more developed countries. Spending on unemployment insurance for LICs and LMICs should not necessarily be zero, but the relative size of the spending should be lower than for ALMPs and EI programs. It is important to create the institutional capacity needed to deploy unemployment insurance programs once the country is further along in its structural transformation, with more people in formal-wage-employment. To successfully implement unemployment insurance, systems and institutions need to be in place. These include, for example, proper delivery mechanisms to provide payments to the unemployed, monitoring tools that allow governments to see if a person has found a job and must no longer receive payments, institutional coordination for budget management, and correct design to align incentives so that the unemployed find jobs quickly. Countries at the lower end of the income distribution should invest in the creation of these systems. EI programs address barriers to access to capital, technology, and knowledge that prevent the most vulnerable and marginalized segments of the population from engaging in productive self-employment. Unlike traditional unemployment assistance programs, EI 16 initiatives are designed to support individuals and communities facing extreme poverty and exclusion. These programs offer a comprehensive approach, aimed at facilitating the transition of those living in extreme poverty to more secure livelihoods. They typically involve offering productive assets, such as livestock or supplies for small-scale trading, along with technical skills training to effectively manage these assets. They may also include support for basic consumption needs, establishment of savings mechanisms, and initiatives facilitating access to health and education. Investment in these types of programs in LICs makes the most sense, given the prevalence of families who depend on insecure and fragile livelihoods, like domestic work and casual farming. Empirical evidence suggests that EI programs enable the poorest to shift out of farm labor to running small businesses, resulting in increased earnings (Bandiera et al. 2017; Banerjee et al. 2021). Box 3.1. Innovations in Labor Market Programs: Global Skills Partnerships Managed migration has the potential to be a driver of growth for LMICs. Migration leads to significant wage increases for most migrants whose skills and attributes align well with the needs of the destination society. These gains from wage differentials are shared with families and communities in the countries of origin through remittances, which have proven to be a key tool for reducing poverty rates and building more resilient communities. Migration also benefits countries of origin through knowledge transfers from diasporas and returning migrants, which foster entrepreneurship and contribute to the integration of origin countries into global networks and the development of local industries. To reap the benefits of migration, migrants must have the skills and attributes needed to match the demand in destination societies. It is crucial to link labor mobility with skill development in sending countries, ensuring that the skills not only fulfill labor market needs in destination countries but also increase human capital in the countries of origin. In this way, international cooperation can enhance the match while reducing the adverse effects of potential brain drain associated with migration. A potential tool to address this challenge through managed migration is Global Skill Partnerships (GSPs). GSPs are bilateral migration agreements that simultaneously address labor market needs and skills shortages in both destination and origin countries while guaranteeing legal entry pathways for migrants. As part of this framework, the destination country provides the technology and financing to train people in the country of origin with the skills that are in demand in both countries. Some of the trainees migrate while others stay and enter the labor market to meet skill demands at home, covering labor shortages in the destination countries while also ensuring skills development in countries of origin. Examples of such initiatives include the Australia Pacific Training Coalition (APTC) and the Pilot Project Addressing Labor Shortages through Innovative Labor Migration (PALIM) between Morocco and Belgium, which is being succeeded by a new GSP in the information and communications technology (ICT) sector that also includes Tunisia’s Toward a Healthy Approach to Labour Migration Governance and Labour Mobility in North Africa (THAMM). These programs have successfully provided training and opportunities for migrants while addressing labor market needs in various sectors. 17 The intended beneficiaries of labor market programs may also vary by a country’s income level, and better targeting can improve both the efficiency and the effectiveness of programs. The first layer of targeting involves deciding whether labor market programs are targeted to the unemployed, disadvantaged, or vulnerable populations or to recent labor market entrants who may be underemployed and could access better-quality jobs. Even once that decision is made, targeting plays a role in determining which interventions to offer to the self-employed and small and medium enterprises (for example, business training, management training or socioemotional skills), which skills training to offer to which job seekers, and how to allocate capital and to which size of firms. Governments’ capacity to effectively implement and oversee labor market programs is critical for scaling up, adapting to evolving labor markets, and responding rapidly to shocks. In many LMICs, labor market programs are managed in a fragmented manner, with separate units handling assessment, enrollment, provision, and follow-up for separate projects. This fragmentation forces governments to reconstruct delivery chains for every new program, limiting scope and agility. Governments play multiple roles—from regulator to provider— further complicating program administration. To address these constraints, countries should build integrated systems that streamline functions across programs and clarify roles within government sectors, and between public and private sectors. HICs offer a model for this, having invested in delivery systems that incorporate intersectoral coordination, participant profiling, referral mechanisms tailored to individual needs, labor market information systems, quality assurance processes, registries for improved targeting, monitoring and evaluation frameworks, and clear provider mapping.5 These integrated systems enable a coordinated response to labor marketdemands, enhancing adaptability and efficiency beyond what isolated, stand-alone programs can achieve. 5 LMIS refers to the institution(s) that generate information in a granular (but not individual-level) format, which serves to inform different labor market actors (education, training, career guidance, or job search). This is different from jobseeker registries in job matching systems and labor registries, where formal workers are registered by employers. 18 Countries can strengthen their delivery systems by building foundational elements across three layers: program-level efficiency, cross-program coordination, and systemwide governance. At the program level, standardized processes—such as beneficiary enrollment, need assessment, and progress monitoring—allow programs to function independently while also enabling data collection on what works, creating a foundation for future integration across programs. Efficient program-level processes also provide governments with valuable insights into workforce demands and program performance, laying the groundwork for more comprehensive improvements across the labor market system. Enhancing coordination across programs benefits both users and administrators. Shared registries, for example, reduce duplication, enabling governments to consolidate beneficiary data for more targeted policy planning. Additionally, coordinated functions such as beneficiary profiling—which goes beyond eligibility screening to align participants with programs offering the highest returns, and referral systems—ensure that individuals access the right resources at the right time. This level of cross-program coordination simplifies access for users, reduces redundancy, and improves efficiency across government operations. Establishing systemwide governance structures is essential for creating streamlined delivery environments. Centralized regulatory functions, such as setting standards and policies, ensure that all programs operate under a consistent framework, while service delivery responsibilities can be effectively delegated to local governments, private entities, or specialized providers. This division fosters accountability, leverages institutional strengths, and enhances system adaptability to labor market changes while minimizing redundancy. By distinguishing between regulatory oversight and service provision, countries can create more resilient labor market delivery systems, equipped to respond to both immediate and long- term challenges. Moreover, these systems must allow programs to adjust during periods of shocks and facilitate the creation of new ones when necessary. Impending crises like fragility, conflict, and climate disasters underscore the importance of adaptability in program design. The 19 number of people residing in conflict-prone areas has doubled since 2007, and as of 2020, fragile and conflict-affected states harbored more individuals living in extreme poverty (with incomes below US$1.90 a day) than all other nations combined (Corral et al. 2020). Furthermore, without a shift in current practices, climate change is projected to force 132 million people into poverty over the next decade, undoing hard-won development gains (Jafino et al. 2020). By 2050, up to 216 million people are anticipated to migrate within their own countries due to diminishing water resources, declining crop yields, and the impacts of rising sea levels and storm surges (Clement et al. 2021). These formidable challenges underscore the imperative for labor market programs capable of responding swiftly during crises, facilitated by robust information and delivery systems that enable flexible adjustments in coverage and benefits, akin to the adaptive measures taken during the COVID-19 pandemic. IV. Current State of Spending on LM Programs Does Not Fully Address the Constraints Faced Using data from ASPIRE, this section examines global trends in labor market programs. The conceptual framework presented in this paper outlined the type of programs expected in countries at different income levels. Using ASPIRE we assess how well current programs align with these constraints. ASPIRE collects data on program expenditure and the number of beneficiaries, based on administrative records of national or federal programs aimed at preventing, managing, and overcoming situations that negatively affect people’s well-being. The sample of labor market programs comprises 458 supply-side labor market programs designed to promote employment, guarantee the efficient operation of labor markets, and protect workers across 76 countries (11 LICs, 21 LMICs, 31 UMICs, and 13 HICs) in 6 regions (19 in Sub-Saharan Africa, 8 in East Asia and the Pacific, 23 in Europe and Central Asia, 12 in Latin America and the Caribbean, 8 in the Middle East and North Africa, and 6 in South Asia) during 2017–2022 (See Appendix Table A.1). ASPIRE collects data on ALMPs and unemployment insurance but not on EI programs. 20 Figure 4.1: The Number of Programs by Country Type and Year Source: World Bank, ASPIRE administrative data - https://www.worldbank.org/aspire. In the ASPIRE sample, most countries have implemented at least one labor market program, irrespective of income level or geographical region. As long-term trends—such as demographic shifts, technological change, climate change, and global integration—reshape the labor market, workers and job seekers can expect more frequent and potentially more challenging job transitions, both within their home country and across borders. Globally, labor market programs are crucial in assisting working-age individuals, particularly youth and vulnerable populations, by facilitating better employment prospects and enhancing productivity. ALMPs are the most common type of labor market policy, with over 80 percent of all the countries in our sample offering at least one of these programs. ALMPs promote opportunity by connecting people to more productive employment by empowering them with new and improved skills, helping them find employment, and increasing their potential earnings through the following programs: (1) labor market services, including intermediation (public employment services), such as counseling, placement assistance, job matching, labor exchanges, and other related services; (2) training, including vocational training, cash for training, workplace training, and others; (3) entrepreneurship support, involving programs that promote entrepreneurship by encouraging the unemployed and 21 target groups to start their own businesses or become self-employed; (4) wage subsidies, comprising employment incentives aimed at covering a proportion of workers’ salaries for a limited period; (5) other ALMPs, such as employment measures for persons with disabilities, job rotation, and job sharing, among others. Additionally, and for this paper, public work programs, typically classified as social assistance programs, are also included as ALMPs. These programs provide cash transfers or food in return for short-term employment at low wages on labor-intensive projects (road construction and maintenance, irrigation infrastructure, reforestation, and soil conservation). When properly designed, they can support labor market activation and serve as stepping stones into longer-term employment, by providing skills and labor market experience to traditionally excluded populations. While ALMPs can significantly contribute to addressing labor market challenges, their scope and effectiveness depend on the presence of delivery systems that can support their implementation. HICs and UMICs place great emphasis on ALMP system-building initiatives, whereas the capacity for systematically delivering ALMPs is often lower in LICs. Consequently, LICs tend to favor stand-alone and productive economic inclusion (PEI) initiatives, which offer a more feasible strategy for addressing pressing employment needs. Nonetheless, PEI programs in LICs are typically centered around agricultural employment, suggesting opportunities for diversification into urban areas and deepening sectoral approaches. Unemployment insurance, often referred to as a passive labor market program, is virtually absent in LICs and increases in importance as income levels rise. ASPIRE disaggregates unemployment insurance into (1) contributory unemployment benefits, which cover workers satisfying the criteria for membership in an unemployment insurance scheme, and (2) non-contributory unemployment assistance, which is intended for workers who either fail to satisfy the criteria for membership in an unemployment insurance scheme or who have exceeded the duration of payments to unemployment insurance benefit. Unemployment insurance is typically absent in LICs, as these programs require more complex delivery systems, and a large proportion of workers are in the informal economy, making it difficult to verify unemployment status. 22 While labor market programs are in place almost everywhere, the spending in these programs is extremely low. As a proportion of total spending on social protection programs, labor market programs account for less than 5 percent, and no region spends more than 0.4 percent of its GDP on these programs. Spending on labor market programs increases with income level and varies by region. Latin America and the Caribbean (0.27 percent) and Europe and Central Asia (0.28 percent) are the two regions that spend the most as a percentage of GDP toon these programs. These figures most likely underestimate the total resources that go into labor market programs, as the ASPIRE data comprise only national-level programs. Programs financed at the subnational level or those both financed and implemented by donor funds or the private sector are not included. In LICs and MICs, the financing of ALMPs is a crucial consideration, particularly given the high levels of informality that characterize their labor markets. While it is often unclear whether informal workers benefit directly from these programs, the extent of their participation is limited in many cases due to eligibility criteria favoring formal sector workers. In terms of financing, ALMPs are generally funded through a combination of public resources. In many countries, these programs rely on general government revenues, ensuring broader coverage. However, in others, they are financed through payroll contributions from formal sector workers and employers, which can restrict access for informal workers. This duality highlights the challenge of ensuring sustainable financing while expanding access to those most in need, including workers outside the formal economy. As labor markets evolve, governments will need to explore more inclusive financing models that enable ALMPs to address the realities of the informal sector effectively. 23 Figure 4.2: Spending on Labor Market Programs is Only Correlated with Unemployment in UMICs Panel A Panel B Panel C Panel D Source: World Bank, ASPIRE administrative data - https://www.worldbank.org/aspire. Note: N = l; h corresponds to the number of countries classified as low and high unemployment \These graphs depict correlations and should not be interpreted as causal effects. There could be reverse causality; HICs that spend more on ALMPs might be better at reducing unemployment rather than HICs with low unemployment being those that spend the most on ALMPs. Spending on labor market programs is only correlated with unemployment levels in upper- middle-income countries (UMICs). Within UMICs, those with higher levels of unemployment also spend significantly more on ALMPs and unemployment insurance schemes. Figure 4.2 illustrates this point by plotting average spending by program type (ALMP and unemployment 24 insurance), normalized by the working-age population, against the unemployment rate for countries across income groups. Low unemployment is defined as an unemployment rate below the mean rate in the overall sample. In ALMPs, the positive correlation between spending and unemployment for UMICs is primarily driven by countries in Latin America and the Caribbean, as well as in the Middle East. For unemployment insurance schemes, Latin America and the Caribbean is the only region where spending is significantly greater among countries with higher unemployment rates. East Asia and the Pacific show almost flat spending profiles across unemployment groups. In all regions, except Europe and Central Asia, spending on ALMPs is positively correlated with self-employment rates. However, the relationship between ALMP spending and self- employment is not uniformly strong, as seen in Figure 4.2 Panel A and B. This highlights the need for a more tailored approach to labor market programs based on specific country contexts. For LICs, where unemployment is less prevalent and much of the workforce engages in self-employment out of necessity, ALMPs should focus on enhancing productivity and income. ALMPs can play a critical role by supporting transitions into more profitable sectors, connecting entrepreneurs to value chains, expanding access to larger markets, and facilitating the growth of small businesses. At the same time, these programs must also prepare non- entrepreneurs to access wage employment opportunities as labor markets evolve. These weak correlations also suggest that governments should identify the beneficiaries of these programs and how programs are targeted. Reaching non-self-employed individuals might be easier, but it is not necessarily where the greatest need lies. 25 Figure 4.3: Spending on LM Programs Is Correlated with Self-Employment in LICs and LMICs Panel A Panel B Panel C Panel D Source: World Bank, ASPIRE administrative data - https://www.worldbank.org/aspire) Note: N = l; h corresponds to the number of countries classified as low and high self-employment, these graphs depict correlations and should not be interpreted as causal effects. Low self-employment is defined as having a self-employment rate below the mean rate in the overall sample. The availability of ALMPs is also clearly associated with country income, with LICs and LMICs having fewer programs. Figure 4.4, Panel A shows the proportion of countries with at least one ALMP by category. As income levels increase, so do the number and types of programs that countries have. In LICs, public works and entrepreneurship support programs are the most prominent. In LMICs, training programs have become more prevalent, indicating a shift 26 toward skills development. Once countries reach UMIC status, they expand their range of programs to include investment in labor market services (for example, public employment services). Finally, HICs have the most diverse array of ALMPs, with wage subsidies being a significant component of their policy toolbox. Interestingly, unemployment insurance— which provides financial support to individuals during periods of unemployment—is completely absent in LICs and becomes more common as income levels increase. This evidence suggests that social safety nets become more robust as countries develop. Beyond social assistance, as countries develop, social safety nets increasingly encompass labor market policies to mitigate the risks associated with the loss of labor income during periods of unemployment or economic instability. Most MICs and virtually all HICs typically have both the resources and, more importantly, the infrastructure to deploy and expand existing labor market programs for workers with diverse needs. Early establishment of labor market delivery systems that can support individuals during transitions in the labor market and in times of vulnerability is crucial for the economic resilience and well-being of workers in LICs. Box 4.1. The State of EI Programs EI programs are used in many regions and contexts to build resilience and create opportunities for poor and vulnerable people. The characteristics of the local economy and target populations shape how programs seek to address the challenges for the poor and vulnerable. Due to limited wage opportunities and high informality in LICs and MICs, 84 percent of programs help develop self-employment opportunities. In MICs, the focus is more on providing wage employment opportunities through ALMPs than in LICs (28 percent in UMICs and 13 percent in LMICs versus 5 percent in LICs), particularly in urban areas. Of the 53 programs identified in the State of Economic Inclusion 2024 that build on social protection programs, only 18 programs focus on wage employment facilitation (World Bank, forthcoming). These programs often focus on skills training, job placement, support, and referrals to increase labor market participation and economic self- sufficiency. Given this trend, there is considerable opportunity to design and scale up programs in urban settings within LICs and LMICs. Urban areas in these contexts often experience high levels of informal employment, where individuals work in unregulated, unstable jobs without formal contracts or benefits. Additionally, urban populations may rely on subsistence agriculture or small-scale businesses, indicating a need for programs that address these specific economic realities (Avalos et al. 2021). Some EI programs are addressing the urban jobs challenge, especially among poor youth and women. For instance, Ethiopia’s Urban Productive Safety Net and Jobs Project (UPSNJP) offers poor urban households temporary income support through public works and facilitates both self-employment (through business capital, training, and other support) and wage employment (through job search assistance and apprenticeships). This program has achieved national scale, reaching over 1.3 million households in 83 cities. 27 This dual approach is particularly effective in contexts with high informality and limited wage job opportunities. Similar programs in other countries are also under way, such as Egypt’s Forsa, Mozambique’s Emprega, and Benin’s Azoli Program. For labor-unconstrained households, this also offers an opportunity to diversify household income, with some members continuing as wage workers while others start household enterprises. While self-employment-focused interventions show promising effects, there is a scarcity of evidence on the impact of EI programs focused on wage employment. In Papua New Guinea, a public works–plus program (Urban Youth Employment Program, UYEP) was able to translate temporary wage employment into formal sector employment. The positive employment impact was achieved by screening candidates on ability and combining job-matching assistance and on-the-job training with wage subsidies to employers (Hoy and Naidoo 2019). There are still critical learning gaps, and systematic evidence is needed to test EI programs focused on wage employment interventions to determine the most effective package for specific contexts and target groups. Green transition also provides an opportunity to invest in EI programming. In light of climate change challenges, many MICs are embracing the green transition agenda. However, this shift introduces several concerns, including potential job losses in specific sectors and the need for substantial workforce retraining and skill development. This transition can leave some people behind—especially those who lack the opportunities and skills to move into new, higher-skilled jobs (IPCC 2022). To address these challenges, EI programs can be tailored to focus on skill development and access to wage employment in green (or climate- neutral) jobs. This can also include job search assistance and placement in sectors or regions where poor and vulnerable groups are likely to be most affected by the transition. By providing targeted support, these programs can help ensure that the green transition does not exacerbate existing inequalities but instead offers new opportunities for those at risk of being left behind. EI programs can play a vital role in addressing social norms and cultural preferences that affect labor force participation, especially for women. Gender roles, childcare responsibilities, and other societal expectations often limit women’s participation in the workforce. Through coaching and mentoring, these programs offer psychosocial support to challenge restrictive social norms and foster community-based networks for women. These networks encourage mutual support, sharing of resources, and collective problem-solving. Additionally, EI programs use community structures throughout the delivery chain to build a sense of ownership and engagement within the community. Establishing community-based childcare facilities or cooperative childcare programs can help alleviate barriers to workforce participation for women. However, spending in labor market programs appears uniformly low and does not seem to correlate with the primary constraints experienced by countries at different stages of development. While the types of existing programs somewhat match the labor market context, as outlined in the conceptual framework presented earlier in this paper, the level of spending on these programs does not align accordingly. For example, LICs allocate similar amounts to wage subsidies as they do to entrepreneurship programs despite the higher prevalence of self-employed individuals and the scarcity of wage opportunities in these contexts. This incongruity highlights the need for a reassessment and realignment of 28 spending priorities to effectively address the specific challenges faced by workers across countries by income level. There is variation in the fragmentation of labor market program delivery across different income groups. LICs exhibit a higher average spending per ALMP than other income groups, indicating a concentration of resources in fewer programs. However, this concentration contrasts with a relatively higher average number of beneficiaries per program, suggesting that while spending is concentrated, the reach of individual programs is broader. As income level rises, there is less spending per program, a lower average number of beneficiaries, and consequently higher spending per beneficiary. This suggests a more targeted approach, with fewer programs catering to a smaller number of beneficiaries, indicating a higher degree of consolidation and efficiency in program delivery. HICs spend more intensively per beneficiary than countries in other income groups. On average, HICs spend approximately US$6,929.54 per beneficiary every year on ALMPs, which is 11 times higher than the spending in LICs. Similarly, they spend US$8,098.31 per beneficiary on unemployment insurance, while LICs allocate zero spending to this type of program. Despite similar spending levels within income groups, there is a notable discrepancy in the number of beneficiaries covered across programs. For all income levels, more people receive ALMP than unemployment insurance benefits, but in richer countries, the difference in coverage is smaller. This data suggests that high-income countries complement unemployment insurance with ALMPs, creating a more comprehensive approach to program design than low- and middle-income countries do. 29 Figure 4.4: Countries at Higher Levels of Income Offer a Wider Range of Labor market Programs, but Spending on Programs Does Not Align with Key Labor Market Constraints Panel A Panel B Source: World Bank, ASPIRE administrative data - https://www.worldbank.org/aspire Note: N corresponds to the number of countries in each income group. For ALMPs: In the sample of 76 countries, 47 countries have training programs, 36 have public works programs, 33 have LM service programs, 18 have wage subsidies, 43 have entrepreneurship support, and 22 have other ALMPs. To calculate spending as a percentage GDP, zero values were included to preserve the consistency of the data with other social protection areas. For unemployment insurance programs: In the sample of 76 countries: 20 have contributory unemployment insurance programs and 7 have noncontributory unemployment insurance programs. To calculate spending as a percentage GDP, zero values were included to preserve the consistency of the data with other social protection areas. EI Program data are missing from these graphs because this is not something that ASPIRE currently captures. The State of Economic Inclusion reports that 14 low-income countries, 23 lower-middle-income countries, and 7 UMICs have EI programs. 30 Figure 4.5: Spending per Beneficiary Increases with Income Level Panel A Panel B Source: World Bank, ASPIRE administrative data - https://www.worldbank.org/aspire) Note: N corresponds to the number of countries in each income group. Average spending per beneficiary was calculated for programs with nonzero data. In total 280 ALMP programs (33 programs in LICs, 87 in LMICs, 88 in UMICs, and 73 in HICs) from 67 countries (11 LICs, 19 LMICs, 24 UMICs, and 13 HICs) and 33 unemployment insurance programs (3 programs in LMICs, 18 in UMICs, and 12 in HICs) from 26 countries (3 LMICs, 15 UMICs, and 8 HICs) were included in the calculations. Each boxplot shows a summary of 5 values. The bottom whisker marks the country with the minimum spending value in each income group. The bottom of the box marks the location of the first quartile or 25th percentile (calculated with the median included). The line within the box refers to the median spending value for the income group. The top of the box represents the third quartile or 75th percentile (also calculated using the median inclusion method). The top whisker signifies the maximum spending value by a country in each income group. The x symbol represents the mean spending for each income group. The boxplots also show internal values and outliers represented by the dots. V. Strong Systems Enable Rapid Responses during Times of Crisis, as Evidenced during the COVID-19 Pandemic Governments use labor market programs not only during periods of stability but also in times of crisis, as demonstrated during the COVID-19 pandemic. The effectiveness and timeliness of government responses to such crises heavily depend on the labor market delivery systems in place. These systems should be capable of swiftly assessing the situation, generating relevant data to identify emerging constraints, and deploying appropriate programs. For instance, during the COVID-19 pandemic, it was crucial to support existing employer-employee matches to prevent job losses and maintain economic stability. Additionally, robust registries and communication mechanisms are essential during a crisis to 31 target individuals and firms that require support. Thus, investing in adaptable and resilient labor market delivery systems is crucial for effectively responding to shocks. Amid the challenges posed by the COVID-19 pandemic, governments worldwide swiftly mobilized their labor market programs to address the effects on employment and livelihoods by using existing programs, expanding them, and creating novel initiatives tailored to the evolving needs of their populations. Notably, UMICs emerged as proactive creators of new programs, leveraging their resources to innovate and adapt to the crisis. Conversely, HICs predominantly opted to capitalize on their robust existing programs, strategically expanding them to meet heightened demand and emerging challenges. In contrast, LICs faced significant hurdles in responding effectively, primarily relying on their limited array of pre-existing programs, which were fewer than those in other countries. This divergence in response strategies underscores the critical importance of programmatic flexibility, resource availability, and institutional capacity in effectively navigating crises and safeguarding the resilience of labor markets. Figure 5.1: UMICs and HICs Expanded and Created New LM Programs During COVID-19 While LICs and LMICs Maintained Existing Programs Source: World Bank, ASPIRE administrative data - https://www.worldbank.org/aspire Note: N corresponds to the number of countries in each income group A total of 403 programs (28 in LICs, 107 in LMICs, 176 in UMICs, and 92 in HICs) from 2020–2021 were considered. In total, 272 programs remained the same (21 in LICs, 83 in LMICs, 105 in UMICs, and 63 in HICs) and 131 programs were either expanded or created. There are 58 expanded programs (4 in LICs, 9 in LMICs, 26 in UMICs, and 19 in HICs) and 73 newly created programs (3 in LICs, 15 in LMICs, 45 in UMICs, and 10 in HICs). Compared to the results 32 in Kamran et al. (2023) we observe less creation of new programs. This is because the universe of programs included in their study was larger, as they included not only programs but also labor policies, as well as small temporary initiatives. The ASPIRE data, by contrast, is a panel data set that focuses on large programs. During the COVID-19 pandemic, spending primarily increased in countries where systems were already in place. However, the growth in spending on labor market programs was not uniform across all countries, and most of the growth occurred in UMICs and HICs. Countries with established programs and systems were able to leverage them effectively during the COVID-19 response, facilitating a quicker recovery. For instance, countries with unemployment insurance made adjustments in response to the immediate shock by providing support to the unemployed, whereas those without such systems faced challenges in addressing inactivity. In the recovery phase, inactive individuals were more likely to rejoin the labor force as informal workers than the unemployed. As a result, informality expanded faster in countries without unemployment insurance. In LICs, the increase in spending was almost negligible, with most of the new spending allocated to entrepreneurship support programs. However, funneling funds into entrepreneurship support during a crisis like the pandemic, which was characterized by both supply and demand shocks, was inefficient. At a time when businesses experienced decreased demand and households suffered income reductions due to business closures and reduced face-to-face interactions, fostering new businesses was probably an ineffective measure. Despite this, governments expanded these services largely because the programs were already established; hence, scaling up was easier. LMICs saw growth in entrepreneurship programs and public works, but this was more than offset by a reduction in spending on training programs. The decrease in investment in training programs can be attributed to several factors, including a decline in wage employment opportunities, restrictions on face-to-face interactions, and challenges in transitioning to virtual training courses. These circumstances underscore the need for adaptive and innovative approaches to address the evolving needs of labor markets in times of crisis. 33 Figure 5.2: Increased Spending Occurred Primarily in Existing Program Types and Decreased After the COVID-19 Pandemic Panel A Panel B Source: World Bank, ASPIRE administrative data - https://www.worldbank.org/aspire. Note: In 2020, wage subsidies for HICs in ALMPs made up 0.63 percent of GDP. Data are not shown on the graph to allow for comparison between income groups. To calculate spending as a percentage of GDP, zero values were included to preserve the consistency of the data with other social protection areas. The number of countries included in the analysis varies by year. In total, 73 countries were considered in 2017 (10 LICs, 21 LMICs, 30 UMICs, and 12 HICs), 76 countries from 2018–2020 (11 LICs, 21 LMICs, 31 UMICs, and 13 HICs), and 71 in 2021 (8 LICs, 21 LMICs, 29 UMICs, and 13 HICs). In UMICs, spending on wage subsidies primarily increased, alongside some growth in unemployment insurance initiatives and entrepreneurship support. Notably, the expansion in wage subsidies outpaced that of unemployment insurance. This emphasis on wage 34 subsidies that UMICs and HICs had during the pandemic aligned with the importance of preserving existing work relationships, particularly to accelerate economic recovery. While receiving either a wage subsidy or an unemployment insurance payment may seem comparable for a worker in the short term, the distinction is stark during the recovery period when COVID-19 restrictions were eased. Maintaining continuity in employment facilitates the transition back into economic activity, compared to the challenges of seeking new employment. HICs exhibited the most substantial increases in spending, a trend consistent with their well-developed public employment services and established delivery systems. This surge in expenditure encompassed both ALMPs, such as wage subsidies, and unemployment insurance. Although their pre-pandemic spending levels were low and comparable to other countries, their adeptness at scaling up and leveraging existing programs was better. Notably, expenditure on wage subsidies increased from 0.05 percent of GDP in 2019 to 0.68 percent in 2020, highlighting the responsiveness of these countries’ labor market programs in the face of crisis. The increase in labor market program spending was only temporary, as seen by the reversal in spending in the years following the initial COVID-19 outbreak. This is, in part, explained by a decreased need as in-person activities resumed. Also, countries’ fiscal capacity reduced due to increasing national indebtedness. Regardless of income, all countries reduced their spending on labor market programs. While this was true for social assistance and social insurance programs, the cutback was larger and more rapid than in other categories. VI. Labor Market Programs Going Forward Labor market programs are important components of the broader social protection agenda as they provide economic opportunities to households. By enhancing labor supply through training programs, stimulating labor demand through apprenticeships or wage subsidies, improving employment quality, and increasing earnings for the self-employed, these programs significantly contribute to the economic well-being of individuals and families. 35 During crises, labor market programs are a crucial government tool to support households through workers. As evidenced during the COVID-19 pandemic, countries that relied on wage subsidies and extended unemployment insurance helped households recover faster by preserving existing job matches in the formal sector. Beyond immediate crisis response, these programs have also helped workers build resilience and adapt to long-term changes, such as those driven by climate change and technological advancements, by providing the necessary skills for new employment opportunities. This paper emphasizes the importance of a cohesive strategy in crafting a country’s labor market program portfolio. First, such strategies should be grounded in addressing prevalent market failures and the core constraints confronting labor markets. For instance, implementing wage subsidies in economies dominated by informal employment may have a limited effect, primarily benefiting only a fraction of the workforce. Therefore, it is imperative for developing countries to tailor their labor market interventions to suit their unique structural shifts and emerging trends, rather than adopting standardized models from more advanced economies. Adopting a comprehensive approach that concurrently tackles multiple constraints is essential. Given the unique and multifaceted nature of labor market challenges, customized and comprehensive approaches tend to outperform interventions targeting specific constraints in isolation. Program design should focus on addressing barriers from both the supply and the demand sides to create more effective programs. In LICs, an effective strategy for this has been the fast expansion of EI programs; in HICs it can be seen through the multiple efforts to combine unemployment insurance benefits with ALMPs to facilitate skills development and transition back to work. Robust delivery systems for labor market programs are paramount. Not only do they ensure more efficient resource allocation, but they also foster a dynamic system capable of swift responses during crises. This was particularly evident during the COVID-19 pandemic when countries with more developed systems in place were better able to respond to the shock by adapting and augmenting their labor market program offerings. Moreover, by relying on labor 36 market programs, countries were able to maintain existing job matches and mitigate switches to the informal sector or into poverty. Current spending levels on these programs are often too low. LICs spend 0.15 percent of their GDP, LMICs 0.27 percent, and UMICs 0.28 percent, compared to an average of about 0.5 percent for OECD countries. Labor market programs have the potential to significantly improve employment outcomes and build overall resilience, but spending in these programs is low. More detailed data are needed to inform discussion on the design and implementation of labor market programs. There is a critical need for enhanced data collection and analysis in this domain. Disaggregated data, detailing program targeting, coverage of demographic groups, such as youth and women, involvement of various agencies, and private sector engagement, are indispensable for gauging program effectiveness. Improved data accessibility and analysis can fuel informed policy dialogues, enabling policy makers to make more precise decisions and tailor interventions for maximum impact. 37 References Apella, Ignacio, Rafael Rofman, and Helena Rovner. 2020. Skills and the Labor Market in a New Era: Managing the Impacts of Population Aging and Technological Change in Uruguay. International Development in Focus. Washington, DC: World Bank Group. https://documents1.worldbank.org/curated/en/817841592200831056/pdf/Skills- and-the-Labor-Market-in-a-New-Era-Managing-the-Impacts-of-Population-Aging- and-Technological-Change-in-Uruguay.pdf. Arévalo-Sánchez, Inés, Janet Heisey, Sarang Chaudhary, Timothy Clay, Victoria Strokova, Puja Vasudeva Dutta, and Colin Andrews. 2024. “The State of Economic Inclusion Report 2024: Pathways to Scale.” Washington, DC: World Bank. Avalos, Jorge, Sarang Chaudhury, Timothy Clay, and Puja Vasudeva Dutta. 2021. “A Path to Jobs for the Urban Poor.” PEI In Practice Vol.1. Washington, DC: World Bank. http://hdl.handle.net/10986/36594. Bandiera, Oriana, Robin Burgess, Narayan Das, Selim Gulesci, Imran Rasul, and Munshi Sulaiman. 2017. “Labor Markets and Poverty in Village Economies.” The Quarterly Journal of Economics 132 (2): 811–870. https://doi.org/10.1093/qje/qjx003. Banerjee, Abhijit, Esther Duflo, and Garima Sharma. 2021. “Long-Term Effects of the Targeting the Ultra Poor Program.” American Economic Review: Insights 3 (4): December. Boserup, Ester. 1970. “Women’s Role in Economic Development.” New York: St. Martin’s Press. Carranza, Eliana, and David McKenzie. 2024. “Job Training and Job Search Assistance Policies in Developing Countries.” Journal of Economic Perspectives 38 (1): Winter. Clement, Viviane Wei Chen, Kanta K. Rigaud, Alex de Sherbinin, Bryan R Jones, Susana Adamo, Jacob Schewe, Nian Sadiq, and Elham Shirin Shabahat. 2021. “Groundswell Part II : Acting on Internal Climate Migration.” Vol. 1. Washington, D.C.: World Bank Group. 38 Corral, Paul, Alexander Irwin, Nandini Krishnan, Daniel Gerszon Mahler, and Tara Vishwanath. . 2020. Fragility and Conflict: On the Front Lines of the Fight against Poverty. Washington, DC: World Bank. Goldin, Claudia. 1995. “The U-Shaped Female Labor Force Function in Economic Development and Economic History.” In Women’s Human Capital and Economic Development, 61– 90. edited by T. P. Schultz. Chicago: University of Chicago Press. Heath, Rachel, and Seema Jayachandran. 2017. “The Causes and Consequences of Increased Female Education and Labor Force Participation in Developing Countries.” National Bureau of Economic Research Working Paper 22766. Hoy, Christopher, and Darian Naidoo. 2019. “The Marginal Benefit of an Active Labor Market Program Relative to a Public Works Program: Evidence from Papua New Guinea.” IZA Journal of Development and Migration 10. https://sciendo.com/article/10.2478/izajodm-2019-0003 International Labour Organization. 2022. ILO Modelled Estimates (ILOEST). Geneva, Switzerland: International Labour Organization. IPCC (Intergovernmental Panel on Climate Change). 2022. “Summary for Policymakers.” In Climate Change 2022: Impacts, Adaptation, and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, edited by H. O. Pörtner, D.C. Roberts, M. Tignor, E.S. Poloczanska, K. Mintenbeck, A. Alegría, M. Craig, S. Langsdorf, S. Löschke, V. Möller, A. Okem, and B. Rama. Cambridge: Cambridge University Press. Jafino, Bramka Arga, Brian Walsh, Julie Rozenberg, and Stephane Hallegatte. 2020. “Revised Estimates of the Impact of Climate Change on Extreme Poverty by 2030.” In Poverty and Shared Prosperity. Washington, DC: World Bank Group. Klasen, Stephan. 2019. “What Explains Uneven Female Labor Force Participation Levels and Trends in Developing Countries?” The World Bank Research Observer, 34(2), 161–197. 39 Liepmann, Hannah, and Clemente Pignatti. 2024. “Welfare Effects of Unemployment Benefits When Informality is High.” Journal of Public Economics 229 (C). McKenzie, David. 2017. “Identifying and Spurring High-Growth Entrepreneurship: Experimental Evidence from a Business Plan Competition.” American Economic Review 107 (8): August. McMillan, Margaret S., and Dani Rodrik. 2011. “Globalization, Structural Change and Productivity Growth.” National Bureau of Economic Research Working Paper 17143. Merotto, Dino, Michael Weber, and Reyes Aterido. 2018. “Pathways to Better Jobs in IDA Countries: Findings from Jobs Diagnostics.” Jobs Series No. 14. Washington, DC: World Bank. http://hdl.handle.net/10986/30594. Nekoei, Arash, and Andrea Weber. 2017. “Does Extending Unemployment Benefits Improve Job Quality?” American Economic Review 107 (2): 527–561. Palacios, Robert, and David A. Robalino. 2020. “Integrating Social Insurance and Social Assistance Programs for the Future World of Labor.” IZA Discussion Paper No. 13258. https://econpapers.repec.org/paper/izaizadps/dp13258.htm. Prescott, Edward C. 1998. “Lawrence R. Klein Lecture 1997: Needed: A Theory of Total Factor Productivity.” International Economic Review 39(3): 525–51. Rofman, Rafael, and Ignacio Apella. 2020. “When We’re Sixty-Four: Opportunities and Challenges for Public Policies in a Population-Aging Context in Latin America.” Washington, DC: World Bank. Romer, Paul M. 1993. “Idea Gaps and Object Gaps in Economic Development.” Journal of Monetary Economics, 32(3): 543–573. Rothstein, Jesse. “Unemployment Insurance and Job Search in the Great Recession.” National Bureau of Economic Research Working Paper 17534. 40 Song, Jae, Joyce Manchester, and Till von Wachter. 2011. “Trends in Employment and Earnings of Allowed and Rejected Applicants to the Social Security Disability Insurance Program.” American Economic Review 101 (7): December. World Bank. 2012. “World Development Report 2013: Jobs.” Washington, DC: World Bank Group. http://hdl.handle.net/10986/11843. World Bank Group. 2022a. “Charting a Course Towards Universal Social Protection: Resilience, Equity, and Opportunity for All.” Washington, DC: World Bank. World Bank Group. 2022b. World Development Indicators. Washington, DC: World Bank Group. https://databank.worldbank.org/source/world-development-indicators. World Bank. 2023. “Skills and Workforce Development.” worldbank.org, October 11, 2023. World Bank. 2025. “State of Social Protection Report 2025: The 2-Billion-Person Challenge.” State of Social Protection Report. Washington, DC: World Bank. World Bank. Forthcoming. “Jobs for Development: Facts and a Framework for Policy.” Washington, DC: World Bank. 41 Appendix Table A.1. Number of Labor Market Programs, Expenditures, and Beneficiaries by Country Income Group, ASPIRE Administrative Data, 2017–2022 Source: World Bank, ASPIRE administrative data - https://www.worldbank.org/aspire. Note: ASPIRE data includes information on 458 active programs in 76 countries between 2017 and 2022. 42 Social Protection & Jobs Discussion Paper Series Titles 2025 No. Title 2511 State of Social Protection Report 2025: The 2-Billion-Person Challenge. Background Paper #4: Optimizing Labor Market Programs and Strengthening Delivery Systems for Impact and Scale 2510 State of Social Protection Report 2025: The 2-Billion-Person Challenge. Background Paper #3: Wake-Up Call for Social Assistance? The Unfinished Mission to Reach the Poor and Beyond 2509 State of Social Protection Report 2025: The 2-Billion-Person Challenge. Background Paper #2: Adaptive Social Protection Agenda: Lessons from Responses to COVID-19 Shock 2508 State of Social Protection Report 2025: The 2-Billion-Person Challenge. Background Paper #1: Mind the Gap: Coverage, Adequacy and Financing Gaps in Social Protection for the Extreme Poor and the Poorest Quintile 2507 Service Integration and Case Management for People on the Move: A Review of Selected International Practices 2506 Impact of Climate Change and the Green Transition on Human Capital: A Review of the Evidence from Europe and Central Asia 2505 A slippery slope: the opportunities and risks of digital approaches and technology in Social Protection Systems 2504 De Jure and De Facto Coverage of Parental Benefits in Nepal 2503 Awareness, Access, and Perceptions around Parental benefits among Urban Argentinians 2502 Regulating Markets So More People Find Better Jobs 2501 São Tomé and Príncipe Unpacking Migration Dynamics: Critical Issues and Policy Recommendations To view Social Protection & Jobs Discussion Papers published prior to 2021, please visit www.worldbank.org/sp. ABSTRACT This paper examines the design and implementation of labor market (LM) programs to address employment challenges across diverse economic contexts. LM programs are critical in enhancing employment opportunities, increasing productivity, and fostering resilience among workers. The analysis highlights gaps in program spending and its alignment with labor market needs, particularly in low- and middle-income countries (LMICs). Using global data, the paper emphasizes the importance of tailored interventions, robust delivery systems, and strategic investments to address structural constraints and adapt to crises, such as the COVID-19 pandemic. The paper also includes a framework for aligning LM programs with labor market realities to support inclusive and sustainable economic growth. JEL CODES J08, J48, H53 KEYWORDS Labor market programs, employment policies, public expenditure ABOUT THIS SERIES Social Protection & Jobs Discussion Papers are published to communicate the results of The World Bank’s work to the development community with the least possible delay. This paper therefore has not been prepared in accordance with the procedures appropriate for formally edited texts. For more information, please visit us online at www.worldbank.org/socialprotection