Education level and age-wealth profiles: An empirical investigation

To test some of the implications of the life cycle-permanent income hypothesis we exploit differences in expected retirement replacement rates between lowly and highly educated households. We examine education-specific age-wealth profiles at the household level. Our sample is an unbalanced panel of 17 years (1994-2010) and approximately 2000 households of Dutch data (DNB Household Survey). We find that, even after controlling for permanent income,highly educated households accumulate more wealth during working life than lowly educated households. A possible explanation is the lower retirement replacement rate of the former group. Furthermore, only highly educated households seem to decumulate wealth after retirement.