✺ How we train
Does training young people for work actually work? What the Dominican evidence says
The Dominican Republic has one of the most thoroughly studied youth training programmes in the world. Its results, six years on, explain why our programme has the shape it has.
A fifteen-year experiment
The Dominican Republic has one of the most thoroughly studied youth training programmes in the world, and almost nobody knows it.
Juventud y Empleo, run by the Ministry of Labour with funding from the Inter-American Development Bank, offered out-of-school young people aged 16 to 29 from poor neighbourhoods 150 hours of technical training in a trade, 75 hours of life skills and a three-month internship in a private company.
What was unusual was not the programme but how it was evaluated. Because there were more applicants than places, places were allocated by lottery. That made it possible to compare, over years, those who got in with those who, identical in every other respect, were left out. It is the same method used to test medicines, and in youth training it had rarely been used outside the United States.
What they found
The first results, 10 to 14 months after graduation, were modest: little difference in the likelihood of having a job, and somewhat higher earnings among those who were working.
With more years and another cohort, the picture changed.
+17% formal employment for men
- Participants were more likely to hold a job with health insurance. Monthly earnings among those working rose 7%.
45% less teenage pregnancy
- Five percentage points lower among female participants aged 16 to 19. The authors attribute it to a shift in expectations about their own future.
Better socio-emotional skills
- Leadership, conflict resolution and personal organisation, measured on standardised scales.
Six years on, the effect was growing
- Men who had been trained were 8 points (26%) more likely to hold a job with health insurance. In Santo Domingo, women earned 25% more than the comparison group.
What the programme did not do was create jobs where there were none. In a country with low unemployment and high informality, training did not increase how many people worked. It changed the quality of the work: from informal to formal, from precarious to stable. And the effects were largest where there was real demand for the skills taught.
Source: Card, Ibarrarán, Regalia, Rosas-Shady and Soares, Journal of Labor Economics, 2011; Ibarrarán, Ripani, Taboada, Villa and García, IZA Journal of Labor & Development, 2014; Ibarrarán, Kluve, Ripani and Rosas-Shady, ILR Review, 2019.
What the rest of the world says
A team of researchers from the ILO and the World Bank reviewed, in 2019, the 113 impact evaluations of youth employment programmes available worldwide. Three conclusions.
The average effect is small
- Only a third of the 3,105 results measured were positive and significant. Training is not magic.
It works better in countries like ours
- Programmes deliver more in middle- and low-income countries than in rich ones.
Design matters more than the type of programme
- Those that combine several services, select carefully whom they serve and follow up individually are the ones that work.
Source: Kluve, Puerto, Robalino, Romero, Rother, Stöterau, Weidenkaff and Witte, Do youth employment programs improve labor market outcomes? A quantitative review, World Development, 2019.
What we learned from all this
We did not invent the programme from scratch. Each decision answers something the evidence had already shown.
Train where there is demand
- Las Terrenas has a sector that hires all year round. We teach nine trades that the town's businesses need today.
Soft skills as a phase, not a talk
- Juventud y Empleo devoted a third of its classroom hours to life skills, and that is where its most lasting effect was. We devote a whole phase to them.
The company is the classroom
- Three days a week in a partner company from week five. The internship is where training becomes provable experience.
Individual follow-up
- Groups of no more than 12, one-to-one sessions on the life project and support for six months after graduation.
Measure and publish
- Explicit targets per cohort: 90% graduation, 90% into work, 95% placed in work-study. If we fall short, we will say so.
The evidence also leaves us a warning. A course does not change a town's labour market. What it changes is each young person's position within that market: from outside the door to inside. That is exactly what we measure.