The importance of employer contacts: Evidence based on selection on observables and internal replication

I estimate effects of the labor market training program "Swit" on employment using both register and survey data. Swit was initiated in an attempt to increase the supply of qualified personnel in the IT sector. Based on the register data I find a large positive effect from the Swit on employment as compared to conventional programs directed towards IT. By also using survey information I conclude that the effect was due to increased employer contacts. The result is of interest because of the relatively large effect especially for individuals with traditionally weak positions on the labor market. Furthermore, I methodologically demonstrate how information about the contents of the programs may corroborate findings based on conditional independence assumptions.