Seeing a colleague lose benefits can change coworkers' behaviour to avoid the same.
Every means-tested benefit programme has to decide who is still eligible. Increasingly, that decision is made automatically: a computer compares a household's registered earnings with an income threshold, and benefits are suspended if income is too high.
When economists evaluate these systems, they usually ask one question: How does the sanctioned household respond? Does it work less? Does it hide income? But this overlooks something important. Losing benefits is rarely a private event. People talk to their coworkers.
Uruguay's AFAM-PE programme, a cash transfer that reached 44% of the country's children during our study period, provides a unique opportunity to study these indirect effects. Every month, the government compared beneficiaries' earnings recorded in the social security system with the programme's income threshold. Households whose verifiable per capita income exceeded that threshold could lose their benefits.
Surprisingly, most beneficiaries did not know that income was checked this way. In a follow-up survey, most respondents could describe the programme's school attendance and health requirements, but only 9% knew that eligibility also depended on registered income. For many families, losing benefits was therefore more than a sanction, it revealed a rule they had not known existed.
In our recent Discussion paper, we examined what happened not only to the households that lost benefits but also to their coworkers. By comparing coworkers of beneficiaries whose income fell just above the eligibility threshold with coworkers of beneficiaries whose income fell just below it we were able to isolate the effect of learning about the rule.
The results were striking. Coworkers who were themselves receiving the programme reacted after seeing a colleague lose benefits. Their registered labour earnings fell by 20% relative to comparable coworkers whose colleagues remained eligible.
The response appeared within two months and was still visible a year later. Most of the decline came from workers leaving formal payroll jobs rather than leaving work altogether. Survey evidence suggests that many moved into informal employment instead. By reducing their registered earnings, they also became nine percentage points less likely to lose their own benefits.
The evidence strongly suggests that information spread through the workplace. The effects were strongest in small firms, where employees interact more closely, and among coworkers doing similar jobs. They were almost three times larger when the sanctioned worker had contacted the programme's call center, making it more likely that they understood and shared the reason for losing benefits. By contrast, coworkers who were not receiving the programme showed no similar changes, suggesting that the results were not driven by changes affecting the firms themselves.
These indirect responses have important policy consequences. If we consider only the households that actually lost benefits, the programme generates an estimated 53 cents of social value for every dollar of net public spending. Once we also account for coworkers' responses, that figure falls to 36 cents. In other words, ignoring workplace spillovers overstates the programme's value by about one-third.
This is not an argument against verifying eligibility. Governments need to ensure that public support reaches the households it is intended to help. But our findings suggest that the costs of enforcement have been underestimated because they extend beyond the households directly affected.
A natural response might be to communicate the rules more clearly. Yet this could cut both ways. Better information might reduce unnecessary reactions based on misunderstanding, but it could also make the income threshold more salient and encourage more people to adjust their earnings to remain eligible. Communication is therefore a policy tool in its own right, and its effects should be tested rather than assumed.
The broader lesson extends well beyond Uruguay. Whenever eligibility for social benefits is monitored using payroll records, the consequences of enforcement are likely to spread through conversations at work. Policymakers who focus only on the households directly affected may miss an important part of the picture and underestimate the true cost of enforcing eligibility rules.
© Joan Vilá, Marcelo Bergolo, and Guillermo Cruces
Joan Vilá is Associate professor at the Instituto de Economia (IECON) at the Universidad de La República, Uruguay
Marcelo Bergolo is Professor at the Instituto de Economia (IECON) at the Universidad de La República, Uruguay, and IZA@LISER Research Fellow
Guillermo Cruces is deputy director of the Center for Distributive, Labor and Social Studies (CEDLAS) at the Universidad Nacional de La Plata, Argentina (UNLP), and IZA@LISER Research Fellow
Please note:
We recognize that World of Labour articles may prompt discussion and possibly controversy. Opinion pieces, such as the one above, capture ideas and debates concisely, and anchor them with real-world examples. Opinions stated here do not necessarily reflect those of the LISER.
Related World of Labour content:
https://wol.iza.org/articles/can-cash-transfers-reduce-child-labor by Furio C. Rosati