Building an equitable future? BRAC’s STAR program and young women’s economic empowerment in Bangladesh

ABSTRACT

Equitable futures depend upon better employment outcomes for young women. We use a skills ecosystems approach to explore how youth skills development programs can maximise their impact in highly gendered societies and labour markets. While improving employment outcomes and incomes for young women, BRAC’s STAR program cannot withstand Bangladesh’s deep-rooted and socially restrictive norms and practices. Short-term successes are diluted as family pressures and commitments take precedent over young women’s economic lives. Adding a temporal dimension to the concept of skills ecosystems, these findings highlight that maintaining these impacts requires constant renegotiation and advocacy to challenge the structural obstacles within households and labour markets and within national policies and investments that constrain their longer-term economic empowerment.

KEYWORDS:

• Youth
• labour and livelihoods
• education
• gender and intersectionality
• diversity and inclusion
• South Asia

1. Introduction

Young people find themselves at the forefront of development challenges globally, especially employment and labour-market challenges. Defined for statistical purposes as those between the ages of 15–24 years old – but with important regional definitional differences reflecting the social nature of age categorisations – young people are disproportionately likely to be unemployed and generally display higher poverty rates even when working. Thirteen per cent of young workers are in extreme poverty and a further 17 per cent in moderate poverty (ILO 2020). These figures are indicative of their widespread engagement with informal work, characterised by insecure and unstable incomes, precarious working conditions, and limited social and legal protections (Flynn et al. 2016). Three out of four young workers worldwide were engaged in informal work in 2016 (ILO 2020). The COVID-19 pandemic escalated these challenges, closing educational institutions, disrupting economies around the globe, and dampening hopes of reaping the “demographic dividend” opened by a global population that is the youngest it has ever been (Oosterom 2018).

The youth employment crisis is visible not only in its scale but also in its nature. High levels of under-, unemployment, and working poverty have deep implications at this formative stage in the life-course, having deep, long-lasting repercussions on young lives (Banks 2020). Beyond the day-to-day economic repercussions of insecure livelihoods, an inability to find regular work leaves young men and women struggling to meet social and cultural benchmarks of adulthood, such as completing education, getting married, and establishing their own family. Much literature highlights the “stuckness” of young people; they find themselves in a limbo between youth-hood and adulthood, unable to aspire to, or achieve social or economic mobility. This is linked to poor mental health outcomes and risky behaviours in diverse contexts (Banks and Sulaiman 2012; Mains, Hadley, and Tessema 2013; Scales, Roehlkepartain, and Shramko 2016).

It is no surprise then, that youth employment is a major focus for youth programming. Yet the evidence base supporting traditional supply-side interventions that build young people’s skills for the workforce or entrepreneurship is weak. Meta-evaluations highlight their limited and variable impact (cf. Cheema 2017), contrasting starkly with the popularity and scale of investment in such programs (Sumburg et al. 2020). Recent moves in the direction of a skills ecosystem approach to understanding skills uptake, utilisation, and employment outcomes are therefore promising (cf. Wedekind et al. 2021; Brown 2022).

A skills ecosystem approach situates an emphasis on skills development within the social, spatial, and ecological contexts in which it takes place, recognising the role of multiple and interdependent ecosystems actors in collaborating to achieve better pathways to jobs and livelihoods (Wedekind et al. 2021). It is to this literature that the current paper contributes, exploring how such an approach can help us to understand the impacts and challenges facing one training program to equip young people for the workforce, namely BRAC’s Skills Training for Advancing Resources (STAR) program in Bangladesh. With socially inclusive skills transitions highlighted as a gap in this existing literature (Hodgson and Spurs 2018), our analysis provides important insight into the concept of inclusive skills ecosystems, both in terms of methodologies for supporting the inclusion of young women and other marginalised groups into exclusionary labour markets and training programs and in terms of grappling with the medium- and long-term challenges to sustaining short-term program improvements in social environments in which multiple influences push and pull young women away from the labour market.

Ultimately, while our analysis highlights the short-term positive outcomes in employment and income that young women gain through participation in STAR, these prove difficult to maintain. While STAR’s ecosystem approach addresses the discriminatory patriarchal norms in the household, community, and local labour market that create barriers to entry for skills training for young women and limit pathways to skills utilisation post-training, shifts in these discriminatory norms are hard to maintain. We conclude by reflecting on the ways in which BRAC can bolster local “horizontal” efforts with greater “vertical” investments that foster supportive and inclusive labour-market policies around learning and working that tackle underlying social norms over the longer term.

We first introduce debates around youth employment challenges and relevant programs, looking at supply-side and more integrated approaches to addressing the youth employment challenge. Meta-evaluations of such programs reveal the limited effectiveness of narrow skills-based approaches for improving youth employment outcomes. We introduce the concept of a “skills ecosystem” to understand the environment in which different groups’ skills are recognised and utilised, locally and nationally. Tackling this broader ecosystem is necessary to challenge the social and cultural norms that limit working opportunities for young women. The next section then introduces BRAC’s STAR program, exploring the program’s successes and the challenges it faces to sustaining improvements in young women’s working lives. We finish by highlighting how lessons from a skills ecosystem approach may support BRAC to further acknowledge and dismantle the deep-rooted patriarchal structures that underpin deep gendered inequalities in Bangladesh’s labour market.

2. Youth employment programs: an overview

2.1. Supply-side interventions for youth employment

Youth employment programs can be situated within a broad array of programs for economic inclusion and empowerment. Demographic and urbanisation mega-trends mean that these are increasingly targeted explicitly to young people: the World Bank’s (2021) State of Economic Inclusion highlights urban youth as a particularly important target group (Andrews et al. 2021).

“Supply-side” interventions have been particularly popular, building youth employability through skills development for the workplace or entrepreneurship. Skills development programs can be split into two broad categories: “workforce development” (e.g. apprenticeships, vocational training, or job match support) and “entrepreneurship” (Flynn et al. 2016). These address the skills gap young people face within these two broad sections of the labour market and may be accompanied by additional services, such as youth-tailored finance (Sumburg et al. 2020).

Skills development programs are increasingly seen as critical strategies for promoting inclusive growth and overcoming poverty traps (Allais 2012; Cheema 2017), for expanding educational access and improving young people’s skills and capabilities (Flynn et al. 2016; Oosterom 2018; Brown 2022), and for creating a responsive workforce in the context of rapid economic, social, and ecological change (Brown 2022). These programs have moved to the top of global policy agendas and have been researched to assess their impacts. But this burgeoning evidence base paints a bleak picture about their effectiveness.

Cheema’s (2017) (non-youth-specific) meta-evaluation of economic empowerment programs finds limited empirical evidence of positive outcomes from supply-side interventions. Looking at youth programming, in particular, Oosterom (2018), finds limited evidence that skills attained lead to sustained employment and increased earnings. Kluve et al.’s (2017) systematic analysis paints a slightly optimistic picture. Looking across 107 youth-focused interventions across 31 countries, they show an overall positive impact on employment and earnings for young people. But these positive effects are small, with a lot of variation between programs (Kluve et al. 2017).

Even where evaluations show positive impacts, the benefits are small in comparison with the large sums of money invested by donors, governments, and NGOs globally in these programs. The “displacement effect” of such programs further reduces their economic value. When job creation cannot keep pace with growing youth populations, improvements for some youth come at the cost of young people that do not receive similar training. As Oosterom (2018:, 4) succinctly describes, “Instead of creating new jobs, they move some youth to the front of the employment queue at the expense of others”.

Such critiques of supply-driven approaches to youth employment have led to recognition of the need for i) broader integrated programs for youth employment and ii) bringing demand- and supply-side interventions together.

2.2. Multi-dimensional programs for youth empowerment

Flynn et al. (2016) highlight that workforce development and entrepreneurship programs sometimes integrate social components into their design. The inclusion of “life skills” may seem out-of-place in labour-market oriented interventions. But economic and social vulnerabilities are so deeply interconnected that they are often inseparable in young lives. As such, integrated programs are evidenced to be more effective than narrow skills-focused interventions, offering greater promise for economic inclusion and displaying greater impacts on income assets and savings (Andrews et al. 2021).

These programs have a different framing of the youth employment crisis. Programs are not seen only as vehicles of skills transmission that provide a means to reach the “end” goal of employment. Instead, integrated programs situate young people’s inability to access decent work and incomes within their broader social and economic vulnerability. To address these interlinked challenges, they combine multiple components to build social and economic skills and capabilities simultaneously.

While the concept of economic empowerment is closely aligned with gender, scholars argue that it should be applied to any group subject to inequalities in access to the labour market based on social attributes such as age, gender, class, disability, race, ethnicity, and caste (Cheema 2017; Oosterom 2018). Intersectionality puts certain groups at greater disadvantage where these inequalities combine. Young people from informal settlements may be subject to widespread discrimination, for example, and gender layers an additional element of exclusion here for young women. In the context of such inequalities, work outcomes can never be improved through skills development alone.

In Cheema’s (2017) meta-analysis of economic empowerment programs, one of the only programs to reveal positive social and economic effects was another flagship program of BRAC, the Empowerment and Livelihoods for Adolescents (ELA) program in East Africa. ELA combines a safe space for young women to meet, socialise, find mentors and role models, play sports and gain life skills along with a range of economic interventions that include skills training and access to tailored microfinance (Banks 2017). The benefits of this integrated approach on members and the broader youth community is evidenced by randomised-control trials. On the economic side, impacts include higher engagement with income-generating activities and incomes for club members. On the social side, a range of improved social outcomes (e.g. better health knowledge, safer sexual and reproductive health decisions and lower fertility rates) also spill over to non-member young women in the broader community (Bandiera et al. 2020). Four years post-intervention, these economic benefits and many of the social outcomes remain significant (Bandiera et al. 2020). While integrated programs combining life skills and livelihoods training that generate strong peer support networks play an important role in improving social and economic outcomes, it is notable that these programs are tailored more to entrepreneurship than wage employment.

2.3. A “skills ecosystems” approach: bridging supply- and demand-led interventions

Youth employment outcomes do not only depend on young people’s skills, labour-market readiness, or willingness to work. They also depend on the strength and characteristics of the labour market, including the power dynamics and structures within it that dictate how opportunities are distributed. A lack of productive employment opportunities may be central to the youth unemployment crisis, but there are no simple or short-term solutions for structural transformation that could help pave the way to decent work for all – including, and especially for, young people. Policymakers face strong pressure to act and coupled with disappointment in the most common demand-side interventions, this has led to a situation in which stakeholders “continue to promote generally ineffective training and skills interventions and cling to the idea that millions of young people can create their own jobs through entrepreneurship” (Sumburg et al. 2020). This is why supply-side interventions have become so popular (Sumburg et al. 2020).

However, programs must also challenge the demand-side of the youth employment crisis. Addressing the discriminatory and exclusionary power dynamics that distribute work opportunities unequally to and among young people is central here (Oosterom 2018). For young women, gendered norms and practices are key constraints. Many stakeholders recognise the specific vulnerabilities faced by girls and young women and thus focus mainly or exclusively on their specific gendered needs related to skill building. It is less common for programs to address discriminatory hiring practices and the norms and belief that affect young women’s ability to apply skills learned in highly gendered labour markets. The issue of skills and skills development has been largely separated from power, the structuring of labour markets, and the social organisation and regulation of occupations and jobs (Allais 2012).

A useful concept to deepen our understanding here is a skills ecosystems approach that prioritises a nuanced and place-based understanding of the social, political, and institutional environments that constitute the “ecosystems” supporting or constraining demand for skilled labour, skills utilisation, and opportunities for ongoing skill development (cf. Hall and Lansbury 2006; Hodgson and Spurs 2018; Wedekind et al. 2021). A skills ecosystem draws attention to the interdependency of multiple actors and policies in creating and sustaining the conditions under which appropriate skills can be developed and deployed (Hall and Lansbury 2006).

Hodgson and Spurs (2018) differentiate between the “collaborative horizontalities” that constitute networks and collaborations between various actors at the local level and the “facilitating verticalities” that constitute the policies and actors that are intended to support learning, living, and working at regional and/or national levels. Effective leadership is necessary to connect these two dimensions, with Wedekind et al. (2021) highlighting the role of “anchor” institutions in this process. “Flourishing” skills ecosystems occur through long-term efforts to understand and promote how best actors and collaborations can come together across these levels to promote inclusive outcomes (Wedekind et al. 2021).

Brown (2022) extends the concept of the skills ecosystem approach to Global South contexts, drawing particular attention to how power relations structure and govern labour markets in the largely informal and precarious labour markets that shape how skills are developed and deployed. This requires a politicised approach to understand the question of whose skills are developed, utilised, and socially valued in labour markets – and whose are not (Brown 2022). This focus on the inclusivity of outcomes has been otherwise broadly missing in analyses of skills transitions and outcomes (Hodgson and Spurs 2018).

With labour markets informally regulated by social institutions that discriminate against particular groups, different employment outcomes are not only due to differences in training quality that individuals or groups receive. Families, communities, and labour markets are all socially discriminatory sites where norms of gender, age, caste, and ethnicity regulate who has opportunities to acquire, utilise and benefit from skills (Brown 2022). This influences the social, institutional, and infrastructural support that different jobseekers have in building skills and searching for work. Moving towards more inclusive outcomes requires a better understanding of how power structures across these different sites lead to “enabling” skills ecosystems for some young people but “constraining” ones for others. This makes the role of different actors at local, regional and national levels in levelling this playing field an important question (Wedekind et al. 2021). So far, however, our understanding of this political economy of skills utilisation has not translated into action around how these deep-rooted constraints and cultural barriers can be overcome in programming (Brown 2022). Within this context, BRAC’s STAR program offers promising potential.

3. Introducing and situating the STAR program

3.1. Youth in Bangladesh

One-third of Bangladeshis are “young”, and the country nears the mid-point of its predicted “demographic dividend” in which its growing working age population can support strong economic growth with the right policies (BIGD 2018). Yet there are few opportunities outside the informal sector for the two million young Bangladeshis entering the labour force each year (BRAC 2019). In 2013, 87 per cent of employment was informal, rising to 90 per cent for women’s employment (ILO 2017).

The ILO (2017) recommends that Bangladesh implements an employment policy and jobs strategy focusing on enhancing employability through skills development. It particularly emphasises the barriers faced by women, disadvantaged groups, people with disabilities, and ethnic minorities (ILO 2017). Ninety per cent of young Bangladeshis who are not in education, earning, or training (NEET) are young women, for example (BIGD 2018). While 60 per cent of young men report being engaged in income-generating activities, this drops to 25 per cent for young women, who also report lower freedom around occupational choice (BIGD 2018). Much more strategic – and inclusive – investments are necessary to unleash Bangladesh’s demographic dividend.

3.2. The STAR program

Youth programming is a core priority for BRAC in Bangladesh. Its Skills Development Programme (SDP) emerged in the 1990s to support young people who dropped out on completion of BRAC’s primary schooling. In the early 2000s BRAC’s adolescent clubs for girls and young women began to integrate a variety of skills training programs into their design. Through learning from these experiences, the urgency of investing in youth and BRAC’s deep contextual understanding of the labour-market opportunities and barriers faced by underprivileged youth, BRAC was motivated to develop SDP’s current flagship program for youth, Skills Training for Advancing Resources (STAR), in 2012.

Developed in collaboration with the ILO and UNICEF, STAR is a six-month skills development program that has graduated nearly 81,000 apprentices across 17 demand-driven trades. The training builds on the traditional Ustad-Shagred (Master-Apprentice) model, where out-of-school adolescents and youth are placed as apprentices with local businesspersons known as “Master Crafts Persons”. The program has a strong focus on social inclusion and incorporating decent work principles into Bangladesh’s local labour markets. Nearly 60 per cent of graduates are young women, and the program also targets explicitly persons with disabilities, transgender people, and other minority groups.

STAR’s overall objective is closely aligned with supply-side youth employment programs, seeking to “produce a well-trained and empowered manpower among youth and thus enhance employment” (BRAC 2019, 6). Trainees – who are selected along strict eligibility criteria based on age, education, and income¹ – rank their top three trades before being placed with a Master Crafts Person (MCP) to learn their skills over a six-month period. This on-the-job skills training also includes weekly classroom-based theoretical training on their trade of choice and “soft skills” of financial and digital literacy, market assessment, social and environmental awareness, employability, and basic English. Unlike traditional apprenticeships in Bangladesh, MCPs and apprentices are given a monthly allowance, and MCPs receive comprehensive training from BRAC to ensure the quality of training provided. The training is conducted in line with the National Technical and Vocational Qualifications Framework, allowing graduates to receive government certification of their skills after a test.

The following section draws upon a variety of internal sources to present, explore, and understand STAR’s outcomes through a skills ecosystems lens, including program documentation, staff expertise, and a range of qualitative and quantitative program evaluations conducted by BRAC Research and Evaluation Division and the BRAC Institute for Governance and Development (BIGD) at BRAC University. Together, these provide robust evidence to accompany the authors’ grounded experience (BRAC 2017; Das 2021; Jahan 2021). A randomised-control trial comparing outcomes across a large randomly selected sample of “control” and “treatment” youth (those who have received STAR training) has been central to programmatic learning. A 2016 survey provided a baseline understanding of eligible youth’s employment status and experiences, with two post-intervention surveys providing objective evidence of STAR’s short- and longer-term effects (Das 2021).² A 2017 survey compared STAR graduate outcomes with those of eligible young people who did not enrol in the program, six months after completion. A second 2018 survey explored whether improvements relative to non-graduates had been sustained after nearly two years.

A skills ecosystem framework sheds light on those aspects of STAR that underpin its inclusive and positive short-term outcomes. BRAC works at multiple levels to sensitise households, employers, and local labour markets to principles of gender equity and decent work. But it also brings new insight into the program’s longer-term challenges, and we conclude by drawing attention to the additional roles BRAC may consider in order to sustain these short-term impacts. Central here would be to build on STAR’s work in fostering horizontal, local-level collaborations with advocacy work that facilitates structural change across a broader array of actors at the regional and national levels.

4. The STAR experience: can skills development programs build and sustain enabling and inclusive skills ecosystems?

4.1. Improved employment and incomes for young men and women

Das’s (2021) baseline survey highlights the unemployment that characterises young Bangladeshi’s working lives: 72 per cent of young men and 81 per cent of young women were unemployed prior to joining STAR. Under-employment was rife among those reporting work. Eighty-nine and 99 per cent of working young men and women, respectively, devoted under an hour a day to income-earning opportunities.

STAR’s outcomes are evidenced by comparing the outcomes of eligible young people who did and did not enrol in the program. Six months post-intervention, STAR had increased aggregate employment by 31 percentage points and participants earned, on average, 23 per cent more than non-STAR graduates (an additional $6.60/month) (Das 2021).³ Income effects were a result of increased hours worked and a shift away from casual work.

There are distinct gendered impacts in program outcomes. A shift away from casual work led both genders to experience wage improvements. But young men benefited disproportionately from income impacts over the long-term due to the different work opportunities they are prepared for (Das 2021). Young women are more likely to become self-employed, while young men progress into higher-paid forms of wage employment.

A further survey revealed STAR graduates to be more resilient to the COVID-19 pandemic’s economic shocks. While intervention and control groups both suffered sharp income and job losses, STAR graduates fared better along important dimensions. Thirty per cent of STAR-trained women remained employed, for example, in comparison with 20 per cent in the control group (Jahan 2021).

4.2. Towards enabling skills ecosystems: building social capital, networks, and decent work

We can dig below these headline findings by applying a skills ecosystem framework to understand improvements in and challenges to skills uptake and utilisation. On-the-job apprenticeships with Master Craft Persons (MCPs) provide the most direct route to improvements in wage employment. Placements are a chance for apprentices to demonstrate their hardworking nature and motivation alongside developing their skills, improving their likelihood of employment afterwards. 11.5 per cent of graduates found employment with MCPs, in comparison with 2.7 per cent of control group participants (BRAC 2019). STAR’s adoption of decent work principles and practices in turn generates post-program improvements in income. MCPs are trained in decent work principles and regularly monitored to ensure that they provide a decent work environment, including fair wages, to trainees.⁴

The STAR program also builds apprentices’ social networks within local labour markets. Not only do apprenticeships open participants up to the broader network of ties that MCPs hold in the marketplace through their interactions and recommendations, STAR takes apprentices to other markets, giving them an idea of their market value as an employee and preparing them to bargain for better salaries. Such social connections are a critical component of successful job searches in Bangladesh’s urban labour market (Banks 2016).

We must take a gender-sensitive approach in our analysis of how STAR builds social capital and networks, and through this, more enabling skills ecosystems that lead to better skills utilisation for young women and marginalised groups. While STAR invests heavily in advocacy and sensitisation work to challenge Bangladesh’s patriarchal norms, these still limit the long-term sustainability of program outcomes, as we explore next.

4.3. A glass ceiling? Limitations to skills ecosystems investments for young women

Can STAR break the deep-rooted societal, cultural, and religious norms and barriers in Bangladesh that limit young women’s potential and overcome gendered disadvantages in the labour market? It certainly tries to. STAR requires that 60 per cent of apprentices be young women, and the RCT findings reveal important progress that suggest STAR’s approach reduces some of the gendered barriers to skills uptake, utilisation, and employment. Yet barriers remain that prevent some young women from joining STAR and many women graduates from maintaining these achievements over the longer term.

Cultural, social, and religious norms all create barriers to entry for women’s apprenticeships. These operate within households and extended families and from the side of MCPs and market committees. STAR recognises these challenges, and program officials invest significant time trying to overcome them. A social inclusion team encourages eligible young women to join STAR and pairs women participants (for confidence) in the same workplace where possible. Family members are sensitised so that they support their daughters' entry onto the program.

Market Committees are key actors within local marketplaces, influencing the hiring practices of employers in the marketplace. Their wider influence can overshadow BRAC’s efforts even where they have sensitised MCPs to the benefits of supporting women learners. Retention is problematic when a wider marketplace follows traditional norms and practices, even when a particular shop-owner may understand the benefits of training women. While the social inclusion team also conducts advocacy work with marketplace leaders, these norms, beliefs, and practices are deep-rooted and take time to change.

These pressures are particularly strong where young women choose non-conventional trades outside of traditionally “female” spaces of work. While women apprentices may select any trade, young women had little desire to break stereotypes and picked gendered trades, such as tailoring, for their apprenticeships. From 2014 STAR encouraged young women into non-conventional training, investing in advocacy with young women, their families, communities, MCP employers, and marketplace committees to enable this.

Only 42 per cent of young women graduate from non-conventional specialisms such as mechanics, graphic design, and IT; while this is significant in Bangladesh’s patriarchal labour market, it is not as high as BRAC would like (BRAC 2017). Post-training outcomes underpin why this remains a logical choice: young women do better in employment outcomes by sticking to conventional trades. Nearly 70 per cent of young women training in non-conventional trades were unable to find a job after training, in comparison with only 37 per cent of those in conventional trades (BRAC 2017). Young women choosing conventional trades saw their average incomes increase from 4,457 BDT to 5,192 BDT, on average, in the two years following training. Those trained in unconventional trades experienced lower average incomes post-training (4,029 BDT) and saw these decrease over time (to 3,850 BDT) (BRAC 2017).

Alongside presenting entry barriers for young women, patriarchal norms and beliefs create a gravitational pull towards a more traditionally gendered status-quo in the longer term. Employment rates for young women in conventional trades fell from 63 per cent post-training to 40 per cent two years later. This drop was even steeper in non-conventional training, from 30 per cent to 17 per cent (BRAC 2017).

Marriage is a key factor in a constraining skills ecosystem for young women. Program effects are biggest for unmarried women: their employment increases by 43 per cent and these effects persist for three years (Mahmood 2020). But evidence also reveals that marriage pulls young women in the opposite direction, with smaller labour-market outcomes for married women in the longer term compared both to their unmarried counterparts and male graduates (Das 2021).

Marriage introduces a range of additional household responsibilities that are compounded when women have children. It also introduces the influence of husbands and in-laws on decision-making around women’s work and mobility. Nearly half (46 per cent) of unemployed young women blamed marriage or family pressure for their inability to find work following training, increasing to 52 per cent for those trained in non-conventional trades (BRAC 2017). The sway of husbands and family members on decision-making is also evident in the fact that even where graduates find employment, this has no significant positive impact on young women’s decision-making power around income, credit, and savings (Mahmood 2020).

While recognising STAR’s role in improving young women’s employment and income outcomes, we must also acknowledge the constraints on these placed by discriminatory structures that disadvantage young women. These insights introduce a dynamic and temporal dimension to the concept of skills ecosystems. Improvements may be hard fought-for but may not endure without continued advocacy against constraining patriarchal norms and beliefs within households, communities, and labour markets. Social inclusion investments must not be limited to program targeting criteria and implementation but must be accompanied by continued advocacy work at all levels if inclusive outcomes are to be sustained.

BRAC may have made progress on questions of “access” in extending STAR training to large numbers of young women and disadvantaged youth, but the program’s focus on skills development is not sufficient in the long-term. Despite significant investments in inclusion, advocacy, and sensitisation, the initial “girl effect” that results in stronger short-term outcomes for young women is replaced by stronger longer-term outcomes for young men. Pressures from husbands, families and communities lead to young women reorienting their economic futures as subservient to their household and caring responsibilities. Their labour, where it continues, moves back to conventional trades (often home-based self-enterprise) that can fit alongside these pressures.

5. Conclusions

Rapidly expanding youth populations and the global youth employment crisis demand that economic inclusion programs must be targeted towards youth (Andrews et al. 2021). Despite significant investments in supply-side programming that seek to build young men and women’s employability through skills training and education, there is clear evidence that narrow approaches to skills development have limited impact (Cheema 2017; Oosterom 2018). Of significance here is the fact that despite widespread recognition of the informal and deeply social nature of labour markets – and the ways in which this generates inequalities in employment outcomes – youth employment programs have largely neglected to address these challenges in their design. A skills ecosystems approach that analyses how power relations structure and govern labour markets and how this influences the inclusivity and sustainability of employment outcomes adds value to our understanding of programmatic outcomes.

Economic empowerment is at the heart of BRAC’s STAR program. One of BRAC’s clear learnings in its five decades of programming is that to empower girls and women they must have a path to the market. Young women’s social empowerment still requires the economic empowerment necessary to underpin this. It was this understanding that moved BRAC away from its earlier socially focused spaces towards a stronger focus on skills development.

Despite this new focus, STAR offers more than a supply-driven approach to skills development. Central here are its efforts to promote decent work and social inclusion, including integrating young women into the labour market. A skills ecosystems framework reveals the role that these “collaborative horizontalities” (Wedekind et al. 2021) play in overcoming young women’s barriers to skills uptake and utilisation, including the sensitisation of families, communities, and key labour-market stakeholders to this end. These efforts recognise and seek to combat the exclusionary norms and forces that improved outcomes are up against.

Amongst a field of skills development programs for youth offering lacklustre evidence of success, STAR was tested rigorously through a randomised-control trial and shows clear signs of success along important indicators. Yet while promoting young women’s participation in training and improving their employment and income in the short-term, its success is less pronounced in ensuring their long-term economic inclusion. Young men benefit disproportionately in the longer term as young women gravitate from waged to self-employment and/or away from the labour market.

This remaining challenge is a critical part of BRAC’s story and one to lean into rather than shy away from. The norms and traditions of Bangladesh’s heavily patriarchal social and religious environment are diluting STAR’s longer-term impacts. These are upheld through hiring practices in the labour market but also reduce young women’s desire to break boundaries and/or lead to pressure on them from husbands, parents, and in-laws to uphold them. This introduces a temporal dimension to the skills ecosystems concept. Creating enabling environments that promote skills utilisation is not enough: we must also maintain them. In socially restrictive contexts continued investments must be made to challenge these norms and traditions. In contexts in which husbands or in-laws do not want young women to work, or in which market committees and employers do not value young women working, retention will always remain a challenge.

BRAC recognises the need to continue working at these broader levels that place ceilings to young women’s retention and future work outcomes. Working with young women alone is not enough. Tackling the deep-rooted structural obstacles within households, communities, and labour markets that constrain young women’s economic empowerment over the longer term requires equal attention and action. Clear from these findings, too, is that investment in local advocacy work is not sufficient to protect short-term gains. Here, the skills ecosystem framework suggests additional entry points that could complement the work BRAC does in facilitating “collaborative horizontalities” at the local level. This includes to invest equally in “facilitating verticalities” at the regional and national level that support more inclusive policies and enabling environments for young people’s learning and working. Given its place and positioning at the national level (cf. Hossain et al. 2021 for evidence of BRAC’s influence on government policy and spending during the COVID-19 pandemic), BRAC has a strong potential to play the role of an anchor institution in joining up these multi-scalar efforts to foster and facilitate deep discussion that leads to gender policy changes beyond STAR itself.

Despite its explicit investments in the local-level advocacy and sensitisation work necessary to realise its goals of social inclusion for young women and other marginalised groups in Bangladesh’s labour market, STAR’s experiences highlight the deep-rooted structures that continue to limit the potential for young women across Bangladesh. Building enabling skills ecosystems is a long-term project requiring continued effort (Wedekind et al. 2021). Maintaining positive and inclusive impacts requires constant renegotiation, advocacy, and support at multiple levels, asking specifically how best actors and collaborations can come together for the system to flourish. Now is the time for BRAC to ask how it can play this anchor role to strive for more inclusive national policies and investments.

Disclosure statement

No potential conflict of interest was reported by the authors.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Notes

1 These participant selection criteria are based on age (14–18, or 20 for disabled participants), education (participants must have been out of school for at least a year), and income (per capita daily income of families must be less than US$1.56).

2 The study covers 60 BRAC Branch offices across which BRAC selected eligible individuals. BRAC’s Research and Evaluation Division randomly selected about 56 youths from the list of eligible youth in each branch office, randomly selecting half of them to be assigned to the training program. The baseline survey covered both treatment and control individuals. The main paper (Das 2021) provides a detailed balancing test of randomisation; the results show that the randomisation is balanced.

3 While income improvements were sustained, the second-round follow up revealed that there were no significant effects on aggregate employment in the longer term. Non-participants were as likely to be working as STAR graduates two years after program completion (Das 2021).

4 It is beyond the scope of this paper to look simultaneously at STAR’s social inclusion and decent work objectives. Here, we focus primarily on the aspect of social inclusion, in line with the paper’s objectives. But STAR prioritises decent work objectives alongside this, recognising this is critical in the informal labour markets in which most young people are seeking work in Bangladesh. Rahman et al. (2021) explain the bottom-up approach that STAR has taken to fostering decent work objectives amongst its Master Crafts People and highlights the impacts and positive spillovers that have been achieved through these efforts. These include in understanding and implementing decent work principles, ensuring diversity in recruitment and changing mindsets, amongst others. These achievements have fed into better working environments and health and safety records.