Abstract The aim of this study is to assess the impact of various institutional variables on the long-term unemployment rate (LAPU) in the Colombian urban labor market. Vector autoregressive models are estimated using microdata from the Great Integrated Household Survey (GEIH), which has national coverage. Monthly data were analyzed for the 13 major capital cities and their metropolitan areas, as well as for the group of 11 intermediate capital cities, resulting in a total of 24 main urban labor markets in Colombia. The dataset includes unionized individuals, individuals with verbal and written contracts, non-labor income, unemployed individuals with subsidies, and individuals receiving severance payments. The results indicate that the growth in the number of unionized employees and non-labor income contributes to increasing the persistence in the duration of unemployment in Colombia. A key finding is that a positive growth in the ratio – gap between individuals with written contracts versus verbal contracts reduces LAPU. This provides evidence of how reducing information asymmetries in the Colombian labor market can improve labor market outcomes and contribute to medium- and long-term social welfare in Colombian urban centers.