According to Marie Jahoda, paid work, in addition to its manifest function (monetary returns), also fulfills numerous latent functions (e.g., social contacts). Empirical studies have examined the relationship between latent functions and unemployment. However, these studies often face shortcomings, such as small sample sizes, crude definitions of employment status groups, or survey data based on non-random samples. The goal of this paper is to empirically test the key assumptions of Jahoda's model using a representative sample. We analyze 2020 data from the Household Panel Study "Labor Market and Social Security" (PASS, N 10,210 individuals from 7,547 households), a survey investigating labor market and poverty in Germany (Trappmann et al., 2019). Compared to previous studies a large sample allows us a detailed analysis of the presumed effects.