Can a temporary financial boost from the government help people who have been out of work for years finally to secure a stable job? In a study of North Macedonia that I conducted with Alex Armand, Federico Tagliati and Yiming Xia, our results suggest that the answer can be a surprising yes.
By looking at a programme that combined personalised job matching with temporary wage subsidies, we find that helping employers to ‘try out’ vulnerable workers can lead to permanent employment. The research shows that these benefits last for years after the government funding stops, particularly for women and those with no prior work experience (Armand et al, 2026).
Evaluating a subsidised employment programme
The study examines the Subsidized Employment Program in North Macedonia, a country in which nearly 30% of people were unemployed and most of them had been out of work for over a year. The programme had two main features: it matched unemployed people with specific job openings based on their skills; and it paid employers a wage subsidy for the first six months, worth roughly half of the average formal wage. In exchange, businesses had to promise to keep the worker employed for at least one full year – six months under subsidy and six months after it ended.
To measure the programme’s success, we used a lottery system. Out of a group of qualified applicants who had already been matched to a specific vacancy, some were randomly chosen to go to a job interview, while others were not. By following these people for three and a half years using official government employment records, we could see exactly how much the programme affected their working lives.
The results were impressive. People who were invited to an interview were 14 percentage points more likely to have a formal job nearly four years later – equivalent to roughly a 60% increase relative to the control group. Their total time spent in employment increased by a massive 85% compared with those who didn’t get an interview. One striking fact from the study is that the programme paid for itself remarkably quickly: the extra money that workers earned was enough to cover the entire cost of their subsidy in just 20 to 22 months. By the end of the study, the increase in earnings for these workers was roughly double the amount that the government had spent on the programme.
Enabling employers to assess workers’ quality at low risk
These research findings challenge the common belief that wage subsidies don’t work in the long run. In many other countries, similar programmes have failed because they were too bureaucratic or only helped people as long as the government was paying the bill. Our study shows that the ‘secret sauce’ is combining the money with careful matching.
The findings update our understanding of ‘information frictions’ – the idea that employers are often reluctant to hire someone with a long gap in their resume because they don’t know if that person will be a good worker. The subsidy acts as a safety net, allowing businesses to assess a worker’s quality without taking on the full financial risk. This was especially true for women and people who had never held a formal job before: these groups saw the biggest gains in long-term employment because the programme gave them a chance to prove their productivity.
Our research also shows that workers didn’t just get paid: they gained lasting habits and skills. People in the programme reported doing more reading, writing and, to a lesser extent, mathematics at work. We also find improvements in their ‘non-cognitive’ skills, particularly a reduction in neuroticism – a trait linked to anxiety. This is consistent with the idea that having a steady job strengthened workers’ sense of control over their lives, which is likely to have helped them keep their positions or find new ones later on.
The value of personalised matching plus targeted training in high-demand fields
Based on our findings, we suggest that governments in middle-income countries should move away from simply handing out ‘vouchers’ to unemployed people and instead focus on personalised matching and screening. When the government does the hard work of finding the right person for the right job, businesses are much more likely to participate and keep those workers for the long haul.
But our research also highlights a major hurdle for scaling up these programmes. While the programme was very effective for those who participated, we find a significant mismatch between the skills that workers had and those that businesses actually needed. Indeed, only about 30% of applicants could be successfully matched with a job opening.
This tells us that subsidies alone cannot solve unemployment. Future policy actions should combine these programmes with targeted training in high-demand fields to ensure that there are enough qualified workers to fill available vacancies. In addition, because the programme’s success depended on local employment offices being well run, investing in the capacity of these government agencies is likely to be important for the programme to work at a larger scale.
Finally, our study is not designed to isolate the exact channels behind these long-term effects, so the mechanisms we highlight here – reduced information frictions and skill acquisition – are best read as suggestive rather than conclusive. Further research following these workers even longer would help to establish whether the early boost in their careers translates into lasting gains for their families in the years to come.
