Abstract
The ongoing transformation of the labor market driven by demographic and technological changes has a significant impact on both labor supply and skill demand, potentially resulting in skills mismatches. One way to address this is to continuously invest in human capital throughout one's working life – a concept promoted by policymakers worldwide. However, before designing effective programs, it is important to understand how decisions about on-the-job training are formed.
The aim of this project is to improve our understanding of how training decisions are made in German establishments. We develop and test hypotheses on previously underexplored factors that influence both managers' and employees' training investment decisions.
In the first part of our project (AEARCTR-0016440), we targeted decision-makers in German firms. To address the lack of exogenous variation in human capital measures in observational data, we designed a conjoint survey experiment introducing random variation in the characteristics of potential training candidates and training courses.
In this second – complementary – part, we conduct a vignette experiment with employees of the same representative sample of German firms. The experiment is embedded in the IAB Linked-Personnel-Panel (LPP) 2026 wave. For establishments that agree to data linkage (approximately 90%), we can additionally link the survey to administrative records on employees and establishments.
The employee vignette is explicitly designed to build on and complement the employer-side study. It examines training demand from the employee perspective using a harmonized design that mirrors key attributes of the manager vignette, while adding dimensions particularly relevant to employees. Our combined approach brings together information on training supply (employers) and training demand (employees) within the same firms.