AbstractWages are a composite measure of the return to education and the return to work experiences. Work experiences are often defined as workers who gain experiences on‐the‐job. However, work experiences can also be part of a study curriculum in vocational secondary education. We estimate the return to in‐school work experiences by comparing the Heckman selection model and the Rubin matching model. First, we show that students with in‐school work experiences earn +16% more in the first years of labour than their theoretical peers. Second, we indicate that both empirical models do not appropriately deal with censored observations in the presence of an informal market. Including information on a set of censored observations increases the effect to +22%.