Image
Getting young people into work
What one year of implementing FBR@WORK in Tunisia reveals
News |
| by Anis Saidi
FBR@WORK is a results-based financing programme for youth employment in Tunisia, implemented under the leadership of the Ministry of Employment and Vocational Training (MEFP), with support from the Swiss State Secretariat for Economic Affairs (SECO) and the Education Outcomes Fund (EOF), a United Nations fund hosted by UNOPS. The programme aims to support around 1,500 young people aged 18 to 35, unemployed for at least one year, at least 50% of them women, into sustainable employment on permanent contracts (CDI). It relies on three implementing partner consortia (Amideast, EFE/WimbeeTech and IECD Tunisie), responsible for identifying, selecting, preparing and supporting these young people through to hiring and beyond, in coordination with the National Agency for Employment and Self-Employment (ANETI).
The first note in this series argued that it is reductive to attribute youth unemployment to a lack of skills and a need for training alone. For many young people out of work for more than a year, what was missing was not training as such, but a local actor able to identify them, select them, connect them with an employer, train them and support them throughout the transition into work, and beyond.
One year on, with three implementing partner consortia fully operational, that diagnosis still holds. But a year of implementation across very different sectors, regions and partnership models points to two findings: (i) for employers recruiting at scale, sourcing, far more than upskilling, is the support they most need from implementing partners; and (ii) even when a local actor is in place and doing its job well, a significant share of young people do not reach the end of the pathway, either because the promised permanent contract never materialises or because the job does not last.
This is the real subject of this second note. One year of implementation is not enough to establish whether delegating the placement and retention of vulnerable young people to private implementing partners works. It does, however, make it possible to identify precisely at which points, and under which conditions, support for a young person loses its effectiveness, and what this means for anyone designing, implementing or funding comparable programmes.
The five findings below follow the same pattern. The programme started from an assumption; implementation challenged it; partners came to understand why; and they adapted, or are still adapting. That last step is where results-based financing makes the difference: because payment is tied to outcomes rather than to a fixed plan of activities, implementing partners have been able to act on what they were learning as they went.
1. Trained Eye: Employers need workers more than workers need training
FBR@WORK was designed around a skills gap: implementing partners were expected, above all, to train young people to meet the needs of the private sector. For employers recruiting at scale, implementation told a different story. What they needed most was sourcing. Technical training is something they can often provide in-house, once they have the workforce.
This is not a problem employers or ANETI can simply solve on their own. Employers lack the local networks to reach these young people and the volumes involved are far beyond ANETI's usual sourcing capacity. Sourcing is also more than having a pool of candidates: the right profiles have to be identified and mobilised quickly, and with judgement. Today, neither the market nor the public employment service, taken separately, is fully equipped or funded to do this.
Three cases illustrate the need: (i) in automotive textiles, a company in Greater Tunis needed 200 sewing machine operators to meet existing orders, failing which production risked being moved to other group sites elsewhere in the region; (ii) in the automotive wiring harness sector, a company based in Jendouba sought to recruit 700 female operators with the same urgency: without available workers in time, it risked losing a major contract; (iii) an international IT services company is looking to recruit around 1,000 employees on permanent contracts. In all three cases, the limiting factor was the ability to mobilise a large number of suitable candidates quickly, far more than training or support.
The implication is that sourcing should no longer be treated as a by-product of mobilisation work. It could become a service in its own right, which implementing partners develop and offer explicitly to employers with large recruitment needs. For these employers, it may well be the key to placement itself.
2. The Tale of Two Cities: The same model can succeed or fail depending on the local ecosystem around it
Every implementing partner works within the same institutional framework: a signed agreement with ANETI and programme coordination at central level. Local authorities were expected to support the programme, mainly by easing administrative procedures. In practice, their role proved far more decisive.
The same implementing partner, in the automotive wiring harness sector, obtained radically different results in two governorates with exactly the same intervention model. In Jendouba, the Governor, the regional employment directorates, the local employment offices (BETI) and the municipalities worked closely with the partner and the employer. They removed administrative obstacles quickly, but also took part in mobilisation itself: relaying information to local communities, organising open days, joining door-to-door campaigns. The partner led the effort and relied heavily on these local relays to reach young people. Over 140 young people signed a permanent contract meeting the programme's eligibility criteria. The mobilisation also led to additional hires outside the programme's scope, including women over 35. In another governorate, collaboration remained largely limited to meetings, without the same operational involvement on the ground. The partner eventually suspended its intervention there, and no permanent contract has been signed to date.
What made the difference was the real involvement of local authorities on the ground. The employment offices played a decisive role in mobilising candidates directly. The regional employment directorate connected the partner with the Governor and opened the way to coordination with delegations and municipalities. The national framework made this possible, but it did not guarantee it: in the second governorate, the same agreement existed and never translated into the same level of mobilisation.
Local institutional mobilisation is therefore not stakeholder engagement around the programme. It is part of the delivery model. Building on the doors opened at national level, each implementing partner has to build and maintain its own local ecosystem of public partners, governorate by governorate. This ability to activate the right institutional relays may prove to be one of the model's key success factors.
3. Call Me Maybe: Relationships are part of getting young people into work
Pre-employment support was designed mainly as training. Staying in contact with young people along the way was treated as a matter of routine. For the hardest-to-reach young people, it turned out to be much more than that.
Contact after training is needed to organise post-placement follow-up and to collect the evidence required to verify results. Where the relationship with the young person is weak, both break down. One partner that ran an entirely remote training session saw 20% of a cohort drop out, because the young people could not be reached to complete their files.
Trust is what keeps the pathway going. One partner explicitly notes that the time spent with young people during in-person pre-employment training serves, among other things, to build a relationship of trust. That trust makes them easier to reach and more responsive to follow-up once training is over.
Building this relationship is therefore part of the intervention, not a side effect of training. It is a practical condition for placement, for follow-up over time, and for collecting the information needed to validate results. It may also be one of the most important things implementing partners bring. For the most vulnerable young people, building trust takes time and a level of intensity that a conventional employment service, with limited resources, struggles to provide.
4. Deal or No Deal: Shifting rules and employer realities turn recruitment plans into moving targets
Implementing partners sign partnership agreements with employers that specify a number of hires. These commitments were expected to translate broadly into actual recruitment. Reality proved much messier. The gap between what employers commit to and what they deliver at the point of hiring is the most recurrent, and the most consistently underestimated, gap in the programme.
It does not concern all partner companies, but it is documented and quantified in several cases: (i) in renewable energy, an agreement provided for 40 hires of solar panel installers, and only 9 materialised; (ii) in hospitality, a hotel committed to 200 hires across several positions (kitchen assistants, restaurant and bar staff, room attendants). This fell to 40, then to only 8 actual hires. The candidates had already completed a two-week work placement, then waited about three weeks without news, long enough for several of them to drop out; (iii) in real estate, an agreement for 100 permanent hires was signed at national level with a franchise group. Recruitment was then carried out by its local agencies, which were not directly bound by the agreement and offered subsidised contracts (CIVP) instead. These contracts are not accepted under FBR@WORK and give young people less long-term security.
Several forces are at play. The first is regulatory. A labour law reform abolishing fixed-term contracts changed employers' risk calculation throughout the first half of the year, making them more cautious about any permanent commitment while the market adjusted. The second is employer risk management at the level of individual jobs: job titles quietly downgraded to reduce pay, or contracts never shared with the young person. The third is internal: human resources may seek to cap the number of permanent contracts while operational managers want to recruit to keep up with production. None of this means employer interest was feigned. It means that a commitment to hire, however sincere, is not a guarantee, whether it comes from a large group, from a trusted relationship with management, or from a signed agreement.
There is no simple answer yet, and partners keep testing. Approaches tried so far include: (i) citing previous cases of unmet commitments when signing agreements, to alert the employer upfront; (ii) avoiding sectors and employers that rely on seasonal contracts; (iii) targeting, where possible, young people who are no longer eligible for subsidised contracts (CIVP); (iv) mobilising local public actors to put pressure on the employer, with mixed results. In the most difficult cases, some partners stopped working with the employer concerned. None of these approaches has yet proved itself systematically. Because payment depends on results rather than on a fixed activity plan, partners are free to drop what does not work and try something else.
5. In It Together: The first weeks of employment may be where support matters most
Signing a permanent contract was treated as the end point of the intervention. Implementation showed that it is not. Placements can fail in the first weeks after signing, even when sourcing and training went well. These first weeks are a high-risk period, yet too often placement is declared a success at the very moment that risk begins.
Two mechanisms explain why. The first is onboarding. Many workplaces have no structured onboarding: no tutor or named contact is assigned to the young person, who in practice receives contradictory instructions depending on which manager they ask. In the hotel sector, partners also reported difficult relations with existing staff, including misleading instructions given to new recruits that ended in dismissal. The second mechanism is informal trial periods. Some employers have young people work for a month without a contract while they decide whether to hire them. At the end of that month, three outcomes recur: the employer decides not to hire; it offers only a seasonal contract where a permanent one was expected; or it eventually offers the permanent contract, but so late that the young person, discouraged by the wait, has already moved on.
Partners that stay involved after signing get markedly better results than those that consider their job done once the contract is signed. They accompany the young person on their first day at work, act as a named contact, handle complaints, mediate with the employer and keep close follow-up during the first weeks, the most critical period for retention. This support is time-bound, but it is demanding. Negotiating with an employer or resolving a conflict in the workplace is a skill in its own right, one that even a training provider with strong teaching capacity does not necessarily have.
This is an entirely new layer of support that the programme has uncovered, and one that implementing partners can offer alongside sourcing and training. Whether it can be delivered at scale, however, remains an open question.
What this means for the future of employment and training policy
The first note in this series had already shifted the diagnosis: youth unemployment is not only a question of skills; it is above all the absence of a local actor able to identify, select, connect and support young people. This first year of implementation makes it possible to go further, by specifying what, within this local work, really makes the difference. For the companies and sectors observed this year, the content of training does not stand out as the most decisive factor, but two other links in the chain: upstream sourcing, meaning the ability to identify and mobilise the right candidates, particularly for large-scale recruiters; and downstream support once the contract is signed, to secure onboarding and prevent the placement from unravelling in the first weeks. This does not mean training is secondary: it remains necessary, but the large companies involved in the programme this year were often ready to take it on themselves.
For large recruiting companies, training, sourcing and post-placement support are therefore not equivalent: in practice, the last two appear at least as decisive as the first, if not more so.
This hierarchy has a concrete implication for public policy. Article 48 (new) of Decree No. 2019-542 of 28 May 2019, as amended by Decree No. 2023-461 of 5 June 2023, already provides a framework for addressing the skills gap through free access to training. There is, however, no equivalent public policy to date that organises and equips sourcing or post-placement support at the same scale (support here does not mean training, but close follow-up of the young person once in the company: presence during the first days, exchanges with their supervisors, mediation in case of difficulty). Yet this is precisely where, according to this first year, the most frequent breaking points in the pathway to employment are concentrated. This gap deserves to be treated as a genuine subject for reflection for the rest of the programme and, more broadly, for public employment policy.
A second contribution specific to this year’s data: this added value is only fully realised if the implementing partner manages to synchronise with all local public actors. On one side is the administrative chain dedicated to employment: ANETI at central level, the regional employment directorates under the Ministry of Employment and Vocational Training (MEFP), and, closer to the ground, the employment and self-employment offices (BETI), ANETI’s local branches responsible for registering jobseekers, disseminating vacancies and local follow-up. On the other side is the chain of territorial authorities: governorates, delegations and municipalities, whose mobilisation helps remove administrative obstacles and multiplies the reach of sourcing on the ground. In this context, the implementing partner plays an interface role, holding together links (the employer, the young person, the administration) that, left to themselves, do not naturally coordinate. This first year suggests that it is precisely here, in this ability to act as an interface and mobilise the whole local ecosystem, that the success or failure of a pathway to employment is decided.
What this reveals about the results-based financing model
A common thread runs through the five findings of this note: implementing partners did not simply execute an intervention plan fixed in advance. They had to adjust it continuously, whether to strengthen sourcing, tighten post-placement support or rethink how they work with local public actors. This latitude was not a given: it owes much to the results-based financing (RBF) model itself, which ties payment to placement and retention outcomes rather than to the delivery of a predefined activity plan.
For implementing partners themselves, this change of logic is experienced as a change of posture. The director of one implementing partner sums up what RBF has required of his organisation: constantly questioning initial assumptions once they meet reality on the ground. His team expected placement with employers to be the main bottleneck. Experience showed the opposite, and the whole sourcing-training-placement chain had to be revised during the programme: in some sectors suffering from a poor image, the difficulty lay upstream, in mobilising and sourcing young people themselves. The same organisation also had to rebuild its monitoring and evaluation tools to measure each beneficiary’s trajectory precisely, a methodological requirement directly linked to the payment conditions of results-based financing.
A second finding specific to this first year: several partners report that practices tested under FBR@WORK (short training modules, targeted outsourcing of some training, strengthened follow-up methods) are already starting to spread to other programmes, even outside results-based financing. A significant share of the learning in this note would therefore probably not have emerged under a more conventional financing model, built around fixed terms of reference with little flexibility.
Getting young people into work remains complex, and success depends on many factors that vary from one young person to another. There are key success factors, but no universal solution. This is precisely why a model that allows partners to keep learning and adjusting matters.
Part of Work in Progress
