Can a pay-for-performance program help the vulnerable find jobs during a pandemic? Experimental evidence from Empleate in Colombia

During a period of COVID-19-induced job losses and mobility restrictions, the government of Colombia's launched Empléate, an innovative Pay-for-performance (P4P) program that targeted impoverished and vulnerable workers. Empléate operated at a national scale and had a novel financial arrangement: in contrast to traditional programs wherein service providers are remunerated based on their activities, service providers in Empleate only received payments based on successful placement of participants into formal employment. They were also granted premia for sustaining participants in formal jobs months for 3 to 6 months after insertion. This article presents the results of a randomized impact evaluation of Empleate conducted between September 2020 to April 2021. The Intention-to-Treat (ITT) estimates show that program participants were 9% more likely to secure a formal job five to eight months post-treatment. Impacts were larger among men and among individuals with work experience in sectors less affected by the pandemic, with the impacts rising to 22% and 17% respectively. There is no evidence of impacts among women and among individuals without secondary education. These ITT impacts likely underestimate real Average Treatment Effects (ATE) due to issues of imperfect compliance. Complementary analysis from survey data suggests creaming, underscoring the importance of ensuring an adequate allocation of financial risk on P4P contracts. Nevertheless, many design features are promising and positive impacts are noteworthy considering the adverse repercussions in Colombia's formal labor market inflicted by the pandemic.